├── AlignGraph
    ├── AlignGraph
    └── AlignGraph.cpp
├── Eval-AlignGraph
    ├── .nfs00000000000025d800016c95
    ├── Eval-AlignGraph
    └── Eval-AlignGraph.cpp
└── README.md


/AlignGraph/AlignGraph:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/baoe/AlignGraph/1fada4c173f7e2111eeaf6da79e28d7fc32e6df2/AlignGraph/AlignGraph


--------------------------------------------------------------------------------
/AlignGraph/AlignGraph.cpp:
--------------------------------------------------------------------------------
   1 | //**********************************************************************************
   2 | //* Title: AlignGraph: algorithm for secondary de novo genome assembly guided by closely related references
   3 | //* Platform: 64-Bit Linux
   4 | //* Author: Ergude Bao
   5 | //* Affliation: Department of Computer Science & Engineering
   6 | //* University of California, Riverside
   7 | //* Date: 03/24/2011
   8 | //* Copy Right: Artistic License 2.0
   9 | //**********************************************************************************
  10 | 
  11 | #include <cstdlib>
  12 | #include <cstring>
  13 | #include <iostream>
  14 | #include <fstream>
  15 | #include <vector>
  16 | #include <sstream>
  17 | #include <math.h>
  18 | #include <algorithm>
  19 | #include <stdlib.h>
  20 | #include <string>
  21 | #include <time.h>
  22 | using namespace std;
  23 | 
  24 | #define TEST
  25 | #define OPTIMIZATION
  26 | 
  27 | #define MAX 99999
  28 | #ifdef OPTIMIZATION
  29 | #define INIT_CONTIG_THRESHOLD 0.5//0
  30 | #else
  31 | #define INIT_CONTIG_THRESHOLD 0.4
  32 | #endif
  33 | #define CONTIG_THRESHOLD 0.5//0.1
  34 | #define THRESHOLD 0.6//0.2
  35 | #define SD 0//5
  36 | #define SI 0//5
  37 | #define BATCH 1000000// BATCH = 2n
  38 | #define COVERAGE 10
  39 | #define EP 5// 5, 20, 100, max
  40 | #define LARGE_CHUNK 1000000//20000
  41 | #define SMALL_CHUNK 20000
  42 | #define MIN_THRESHOLD 0.1
  43 | 
  44 | typedef struct structSegment
  45 | {
  46 |         unsigned int sourceStart;
  47 |         unsigned int targetStart;
  48 |         unsigned int size;
  49 | } Segment;
  50 | 
  51 | typedef struct contiMerStruct
  52 | {
  53 | 	char nucleotide;
  54 |         unsigned int contigID;
  55 |         unsigned int contigOffset;
  56 |         //unsigned int previousID;
  57 |         //unsigned int previousOffset;
  58 |         //unsigned int previousItem;
  59 |         unsigned int nextID;
  60 |         unsigned int nextOffset;
  61 |         unsigned int nextItem;
  62 | } ContiMer;
  63 | 
  64 | typedef struct nextStruct
  65 | {
  66 |         unsigned int nextID;
  67 |         unsigned int nextOffset;
  68 |         unsigned int nextItem;
  69 | } Next;
  70 | 
  71 | typedef struct previousStruct
  72 | {
  73 |         unsigned int previousID;
  74 |         unsigned int previousOffset;
  75 |         unsigned int previousItem;
  76 | } Previous;
  77 | 
  78 | typedef struct kMerStruct
  79 | {
  80 |         unsigned int traversed;
  81 |         vector<char> s;
  82 |         unsigned int contigID;// omitted, since it is rare that two k-mers at the same genome position with the same contig offset have different contig IDs
  83 |         unsigned int contigOffset;
  84 |         unsigned int contigID0;// omitted with the same reason with above
  85 |         unsigned int contigOffset0;
  86 |         unsigned int chromosomeID0;
  87 |         unsigned int chromosomeOffset0;
  88 |         vector<Next> next;
  89 | //      vector<Previous> previous;
  90 | //      unsigned int nextID;
  91 | //      unsigned int nextOffset;
  92 | //      unsigned int nextItem;
  93 | //      unsigned int previousID;
  94 | //      unsigned int previousOffset;
  95 | //      unsigned int previousItem;
  96 |         int coverage;
  97 | 	int A, C, G, T, N;
  98 | } KMer;
  99 | 
 100 | typedef struct baseStruct
 101 | {
 102 |         vector<KMer> kMer;
 103 |         char nucleotide;
 104 |         vector<ContiMer> contiMer;
 105 | } Base;
 106 | 
 107 | typedef struct positionStruct
 108 | {
 109 |         unsigned int chromosomeID;
 110 |         unsigned int chromosomeOffset;
 111 | } Position;
 112 | 
 113 | typedef struct seqStruct
 114 | {
 115 |         int adjusted;
 116 |         vector<char> nucleotides;
 117 |         vector<vector<Position> > positionSets;
 118 |         vector<int> frSets;
 119 |         int outputted;
 120 |         int ID;
 121 | } Seq;
 122 | 
 123 | typedef struct contigStruct
 124 | {
 125 |         int extended;
 126 | ///     unsigned int contigID;
 127 | ///     unsigned int contigID0;
 128 | ///     unsigned int contigOffset;
 129 | ///     unsigned int contigOffset0;
 130 |         unsigned int startID;
 131 |         unsigned int startOffset;
 132 |         unsigned int endID;
 133 |         unsigned int endOffset;
 134 |         unsigned int startID0;
 135 |         unsigned int startOffset0;
 136 |         unsigned int endID0;
 137 |         unsigned int endOffset0;
 138 |         vector<char> nucleotides;
 139 | } Contig;
 140 | 
 141 | typedef struct contigPositionStruct
 142 | {
 143 |         int targetID;
 144 |         unsigned int sourceStart;
 145 |         unsigned int sourceEnd;
 146 |         unsigned int targetStart;
 147 |         unsigned int targetEnd;
 148 |         int fr;
 149 | } ContigPosition;
 150 | 
 151 | typedef struct contigBaseStruct
 152 | {
 153 |         char base;
 154 |         int coverage;
 155 | } ContigBase;
 156 | 
 157 | typedef struct insertStruct
 158 | {
 159 |         int distanceLow;
 160 |         int distanceHigh;
 161 |         int numChromosomes;
 162 |         int fastMap;
 163 |         int iterativeMap;
 164 | } Insert;
 165 | 
 166 | vector<vector<Base> > genome;
 167 | 
 168 | vector<string> genomeIds;
 169 | 
 170 | vector<Contig> contigs;
 171 | 
 172 | vector<string> contigIds;
 173 | 
 174 | string itoa(int number)
 175 | {
 176 |    stringstream ss;//create a stringstream
 177 |    ss << number;//add number to the stream
 178 |    return ss.str();//return a string with the contents of the stream
 179 | }
 180 | 
 181 | void parseBOWTIE(string buf, unsigned int & targetID, unsigned int & targetStart, unsigned int & targetEnd, unsigned int & targetGap, unsigned int & sourceID, unsigned int & sourceStart, unsigned int & sourceEnd, unsigned int & sourceGap, unsigned int & sourceSize, vector<Segment> & seg, unsigned int & fr)
 182 | {
 183 | 	int item = 0, i, j, j0 = 0, j1 = 0, j2 = 0, j3 = 0, j4 = 0, k, insertion = 0, deletion = 0, total = 0, start = 0, end = 0, tag = 1, IDTag = 0;
 184 | 	char sourceIDBuf[100] = {'\0'}, targetIDBuf[100] = {'\0'}, sourceStartBuf[100] = {'\0'}, targetStartBuf[100] = {'\0'}, CIGARBuf[100] = {'\0'}, frBuf[100] = {'\0'}, realTargetIDBuf[100] = {'\0'};
 185 | 	Segment s;
 186 | 
 187 | 	for(i = 0; i < buf.size(); i ++)
 188 |         {
 189 |                 if(buf[i] == '	')
 190 |                 {
 191 |                         item ++;
 192 |                         continue;
 193 |                 }
 194 | 
 195 | 		if(buf[i] == '\0') break;
 196 | 
 197 | 		if(item == 0)
 198 | 			sourceIDBuf[j3 ++] = buf[i];
 199 | 		if(item == 1)
 200 | 			frBuf[j4 ++] = buf[i];
 201 |                 if(item == 2)
 202 | 		{
 203 | 			if(buf[i] == '*')
 204 | 			{
 205 | 				sourceID = atoi(sourceIDBuf);
 206 | 				fr = ((atoi(frBuf) & 0x00000010) == 0x00000010) ? 1 : 0;
 207 | 				targetID = targetStart = targetEnd = targetGap = sourceStart = sourceEnd = sourceGap = sourceSize = -1;
 208 | 				return;
 209 | 			}
 210 | 			if(buf[i] == '.') IDTag = 1;
 211 | 			if(IDTag == 0) targetIDBuf[j0 ++] = buf[i];
 212 | 		}
 213 | 		if(item == 3)
 214 | 			targetStartBuf[j1 ++] = buf[i];
 215 | 		if(item == 5)
 216 | 			if(buf[i] == '0' || buf[i] == '1' || buf[i] == '2' || buf[i] == '3' || buf[i] == '4' || buf[i] == '5' || buf[i] == '6' || buf[i] == '7' || buf[i] == '8' || buf[i] == '9')
 217 | 			{
 218 | 				CIGARBuf[j2 ++] = buf[i];
 219 | 			}
 220 | 			else if(buf[i] == 'I')
 221 | 			{
 222 | 				insertion = insertion + atoi(CIGARBuf);
 223 | 				total = total + atoi(CIGARBuf);
 224 | 				for(k = 0; k < j2; k ++)
 225 | 					CIGARBuf[k] = '\0';
 226 | 				j2 = 0;
 227 | 			}
 228 | 			else if(buf[i] == 'D')
 229 | 			{
 230 | 				deletion = deletion + atoi(CIGARBuf);
 231 | //				total = total + atoi(CIGARBuf);
 232 | 				for(k = 0; k < j2; k ++)
 233 | 					CIGARBuf[k] = '\0';
 234 | 				j2 = 0;
 235 | 			}
 236 | 			else if(buf[i] == 'M')
 237 | 			{
 238 | 				s.sourceStart = total;
 239 | 				s.targetStart = atoi(targetStartBuf) + total + deletion - start - insertion - 1;// offset is 1-based
 240 | 				s.size = atoi(CIGARBuf);
 241 | 				seg.push_back(s);
 242 | 				total = total + atoi(CIGARBuf);
 243 | 				for(k = 0; k < j2; k ++)
 244 | 					CIGARBuf[k] = '\0';
 245 | 				j2 = tag = 0;
 246 | 			}
 247 | 			else if(buf[i] == 'S' && tag)
 248 | 			{
 249 | 				start = atoi(CIGARBuf);
 250 | 				total = total + atoi(CIGARBuf);
 251 | 				for(k = 0; k < j2; k ++)
 252 | 					CIGARBuf[k] = '\0';
 253 | 				j2 = tag = 0;
 254 | 			}
 255 | 			else if(buf[i] == 'S')
 256 | 			{
 257 | 				end = atoi(CIGARBuf);
 258 | 				total = total + atoi(CIGARBuf);
 259 | 				for(k = 0; k < j2; k ++)
 260 | 					CIGARBuf[k] = '\0';
 261 | 				j2 = 0;
 262 | 			}
 263 | 			else
 264 | 			{
 265 | 				if(buf[i] != '*')
 266 | 				{
 267 | 					cout << "unknown character: " << buf[i] << endl;
 268 | 					exit(-1);
 269 | 				}
 270 | 			}
 271 |         }
 272 | 	sourceID = atoi(sourceIDBuf);
 273 | 	sourceStart = start;
 274 | 	sourceEnd = total - end;
 275 | 	sourceGap = insertion;
 276 | 	sourceSize = total;
 277 | 	if(IDTag == 0)
 278 | 		targetID = targetIDBuf[0] == '*' ? -1 : 0;// atoi(targetIDBuf);
 279 | 	else
 280 | 		targetID = targetIDBuf[0] == '*' ? -1 : atoi(targetIDBuf);
 281 | 	targetStart = atoi(targetStartBuf) - 1;// offset is 1-based
 282 | 	targetEnd = targetStart + total + deletion - insertion;// added: - insertion
 283 | 	targetGap = deletion;
 284 | 	fr = ((atoi(frBuf) & 0x00000010) == 0x00000010) ? 1 : 0;
 285 | }
 286 | 
 287 | void loadGenome(vector<vector<Base> > & genome, int chromosomeID)
 288 | {
 289 | 	vector<Base> chromosome;
 290 | 	string buf;
 291 | 	Base b;
 292 | 	int gp, cp, i;
 293 | 	ifstream g;
 294 | 	string s;
 295 | 
 296 | 	s = "tmp/_genome." + itoa(chromosomeID) + ".fa";
 297 | 	g.open(s.c_str());
 298 | 	if(g.is_open())
 299 | 	{
 300 | 		while(g.good())
 301 | 		{
 302 | 			getline(g, buf);
 303 | 			if(buf[0] == 0) break;
 304 | 
 305 | 			if(buf[0] == '>')
 306 | 				genome.push_back(chromosome);
 307 | 			else
 308 | 				for(i = 0; i < buf.size(); i ++)
 309 | 				{
 310 | 					b.nucleotide = buf[i];
 311 | 					genome[genome.size() - 1].push_back(b);
 312 | 				}
 313 | 		}
 314 | 	}
 315 | 	else
 316 | 	{
 317 | 		cout << "CANNOT OPEN FILE!" << endl;
 318 | 		exit(-1);
 319 | 	}
 320 | }
 321 | 
 322 | void loadSeq(ifstream & in, vector<Seq> & seqs)
 323 | {
 324 |         string buf;
 325 |         int i, sp, ssp;
 326 |         Seq s;
 327 | 
 328 |         s.adjusted = 0;
 329 | 	s.outputted = 0;	
 330 | 
 331 |         if(in.is_open())
 332 |         {
 333 |                 while(in.good())
 334 |                 {
 335 |                         getline(in, buf);
 336 |                         if(buf[0] == 0)
 337 |                                 break;
 338 | 
 339 |                         if(buf[0] == '>')
 340 | 			{
 341 | 				for(i = 0; i < buf.size(); i ++)
 342 | 					if(buf[i] == '.')
 343 | 						break;
 344 | 				s.ID = atoi(buf.substr(i + 1, buf.size()).c_str());
 345 |                                 seqs.push_back(s);
 346 | 			}
 347 |                         else
 348 |                         {
 349 |                                 for(i = 0; i < buf.size(); i ++)
 350 |                                         seqs[seqs.size() - 1].nucleotides.push_back(buf[i]);
 351 |                         }
 352 |                 }
 353 |         }
 354 |         else
 355 |         {
 356 |                 cout << "CANNOT OPEN FILE!" << endl;
 357 |                 exit(-1);
 358 |         }
 359 | }
 360 | 
 361 | int loadSeq(ifstream & in, vector<Seq> & seqs, int & aliStartID, int & seqStartID)
 362 | {
 363 | 	string buf;
 364 | 	int i, sp, ssp, p;
 365 | 	Seq s;
 366 | 
 367 | 	aliStartID = seqStartID + 1;
 368 | 	s.adjusted = 0;
 369 | 	p = 0;
 370 | 	if(in.is_open())
 371 | 	{
 372 | 		while(in.good())
 373 | 		{
 374 | 			getline(in, buf);
 375 | 			if(buf[0] == 0)
 376 | 				break;
 377 | 
 378 | 			if(buf[0] == '>')
 379 | 			{
 380 | 				seqs.push_back(s);
 381 | 			}
 382 | 			else
 383 | 			{
 384 | 				for(i = 0; i < buf.size(); i ++)
 385 | 					seqs[seqs.size() - 1].nucleotides.push_back(buf[i]);
 386 | 
 387 | 				if(p == 1)
 388 | 				{
 389 | 					seqStartID ++;
 390 | 					if((seqStartID + 1) % BATCH == 0) return 0;
 391 | 					p = 0;
 392 | 				}
 393 | 				else
 394 | 					p = 1;
 395 | 			}
 396 | 		}
 397 | 		return 1;
 398 | 	}
 399 | 	else
 400 | 	{
 401 | 		cout << "CANNOT OPEN FILE!" << endl;
 402 | 		exit(-1);
 403 | 	}
 404 | }
 405 | 
 406 | int parseBLAT(string buf, unsigned int & targetID, unsigned int & targetStart, unsigned int & targetEnd, unsigned int & targetGap, unsigned int & sourceID, unsigned int & sourceStart, unsigned int & sourceEnd, unsigned int & sourceGap, unsigned int & sourceSize, vector<Segment> & seg, unsigned int & fr, unsigned int & targetSize)
 407 | {
 408 |         char targetIDBuf[100] = {'\0'}, targetStartBuf[100] = {'\0'}, targetEndBuf[100] = {'\0'}, targetGapBuf[100] = {'\0'}, sourceIDBuf[100] = {'\0'}, sourceStartBuf[100] = {'\0'}, sourceEndBuf[100] = {'\0'}, sourceGapBuf[100] = {'\0'}, sourceSizeBuf[100] = {'\0'}, blockBuf[100] = {'\0'}, targetSizeBuf[100] = {'\0'}, realSourceIDBuf[100] = {'\0'}, realBuf[100] = {'\0'};
 409 |         int item = 0, i, j0 = 0, j1 = 0, j2 = 0, j3 = 0, j4 = 0, j5 = 0, j6 = 0, j7 = 0, j8 = 0, j9 = 0, j, k, sp, j10 = 0, j11 = 0, j12 = 0;
 410 |         Segment s;
 411 | 	fr = -1;
 412 | 
 413 |         seg.clear();
 414 | 
 415 |         for(i = 0; i < buf.size(); i ++)
 416 |         {
 417 |                 if(buf[i] == '	')
 418 |                 {
 419 |                         item ++;
 420 | 			sp = 0;
 421 |                         continue;
 422 |                 }
 423 | 
 424 | 		if(buf[i] == '\0') break;
 425 | 
 426 |                 if(item == 13)
 427 |                         targetIDBuf[j0 ++] = buf[i];
 428 |                 if(item == 15)
 429 |                         targetStartBuf[j1 ++] = buf[i];
 430 |                 if(item == 16)
 431 |                         targetEndBuf[j2 ++] = buf[i];
 432 |                 if(item == 7)
 433 |                         targetGapBuf[j3 ++] = buf[i];
 434 |                 if(item == 9)
 435 |                         sourceIDBuf[j4 ++] = buf[i];
 436 |                 if(item == 11)
 437 |                         sourceStartBuf[j5 ++] = buf[i];
 438 |                 if(item == 12)
 439 |                         sourceEndBuf[j6 ++] = buf[i];
 440 |                 if(item == 5)
 441 |                         sourceGapBuf[j7 ++] = buf[i];
 442 |                 if(item == 10)
 443 |                         sourceSizeBuf[j8 ++] = buf[i];
 444 |                 if(item == 18)
 445 |                 {
 446 |                         if(buf[i] == ',')
 447 |                         {
 448 |                                 s.sourceStart = s.targetStart = -1;
 449 |                                 s.size = atoi(blockBuf);
 450 |                                 seg.push_back(s);
 451 | 
 452 |                                 for(k = 0; k < j9; k ++)
 453 |                                         blockBuf[k] = '\0';
 454 |                                 j9 = 0;
 455 |                         }
 456 |                         else
 457 |                                 blockBuf[j9 ++] = buf[i];
 458 |                 }
 459 |                 if(item == 19)
 460 |                 {
 461 |                         if(buf[i] == ',')
 462 |                         {
 463 |                                 seg[sp ++].sourceStart = atoi(blockBuf);
 464 | 
 465 |                                 for(k = 0; k < j9; k ++)
 466 |                                         blockBuf[k] = '\0';
 467 |                                 j9 = 0;
 468 |                         }
 469 |                         else
 470 |                                 blockBuf[j9 ++] = buf[i];
 471 |                 }
 472 |                 if(item == 20)
 473 |                 {
 474 |                         if(buf[i] == '\0')
 475 |                                 break;
 476 |                         if(buf[i] == ',')
 477 |                         {
 478 |                                 seg[sp ++].targetStart = atoi(blockBuf);
 479 | 
 480 |                                 for(k = 0; k < j9; k ++)
 481 |                                         blockBuf[k] = '\0';
 482 |                                 j9 = 0;
 483 |                         }
 484 |                         else
 485 |                                 blockBuf[j9 ++] = buf[i];
 486 |                 }
 487 |                 if(item == 8 && fr == -1)
 488 |                         fr = buf[i] == '+' ? 0 : 1;
 489 | 		if(item == 14)
 490 |                         targetSizeBuf[j10 ++] = buf[i];
 491 |         }
 492 |         targetID = atoi(targetIDBuf);
 493 |         targetStart = atoi(targetStartBuf);
 494 |         targetEnd = atoi(targetEndBuf);
 495 |         targetGap = atoi(targetGapBuf);
 496 | //      sourceID = atoi(sourceIDBuf);
 497 |         sourceStart = atoi(sourceStartBuf);
 498 |         sourceEnd = atoi(sourceEndBuf);
 499 |         sourceGap = atoi(sourceGapBuf);
 500 | //      sourceSize = atoi(sourceSizeBuf);
 501 | 	targetSize = atoi(targetSizeBuf);
 502 | 
 503 | 	for(i = 0; i < 100; i ++)
 504 | 		if(sourceIDBuf[i] == '.')
 505 | 			break;
 506 | 	if(i < 100)
 507 | 	{
 508 | 		for(j = 0; j < i; j ++)
 509 | 			realSourceIDBuf[j11 ++] = sourceIDBuf[j];
 510 | 		for(j = i + 1; j < 100; j ++)
 511 | 			realBuf[j12 ++] = sourceIDBuf[j];
 512 | 		sourceID = atoi(realSourceIDBuf);
 513 | 		sourceSize = atoi(sourceSizeBuf) ;
 514 | 		return atoi(realBuf);
 515 | 	}
 516 | 	else
 517 | 	{
 518 | 		sourceID = atoi(sourceIDBuf);
 519 | 		sourceSize = atoi(sourceSizeBuf);
 520 | 		return sourceSize;
 521 | 	}	
 522 | }
 523 | 
 524 | void parseDelta(string buf, int & a, int & b, int & c, int & d, int & realA, int & realB)
 525 | {
 526 |         char aBuf[100] = {'\0'}, bBuf[100] = {'\0'}, cBuf[100] = {'\0'}, dBuf[100] = {'\0'}, realABuf[100] = {'\0'}, realBBuf[100] = {'\0'};
 527 |         int item = 0, j0 = 0, j1 = 0, j2 = 0, j3 = 0, rj0 = 0, rj1 = 0, i, aTag, bTag;
 528 | 
 529 |         for(i = 0; i < buf.size(); i ++)
 530 |         {
 531 |                 if(buf[i] == ' ')
 532 |                 {
 533 |                         item ++;
 534 |                         continue;
 535 |                 }
 536 | 
 537 |                 if(buf[i] == '\0') break;
 538 | 
 539 |                 if(item == 0 && buf[i] != '>')
 540 |                         aBuf[j0 ++] = buf[i];
 541 |                 if(item == 1)
 542 |                         bBuf[j1 ++] = buf[i];
 543 |                 if(item == 2)
 544 |                         cBuf[j2 ++] = buf[i];
 545 |                 if(item == 3)
 546 |                         dBuf[j3 ++] = buf[i];
 547 |         }
 548 |         for(i = 0, aTag = 0; i < 100; i ++)
 549 |         {
 550 |                 if(aBuf[i] == '.')
 551 |                 {
 552 |                         aTag = 1;
 553 |                         continue;
 554 |                 }
 555 | 
 556 |                 if(aBuf[i] == '\0') break;
 557 | 
 558 |                 if(aTag == 1)
 559 |                 {
 560 |                         realABuf[rj0 ++] = aBuf[i];
 561 |                         aBuf[i] = '\0';
 562 |                 }
 563 |         }
 564 |         for(i = 0, bTag = 0; i < 100; i ++)
 565 |         {
 566 |                 if(bBuf[i] == '.')
 567 |                 {
 568 |                         bTag = 1;
 569 |                         continue;
 570 |                 }
 571 | 
 572 |                 if(bBuf[i] == '\0') break;
 573 | 
 574 |                 if(bTag == 1)
 575 |                 {
 576 |                         realBBuf[rj0 ++] = bBuf[i];
 577 |                         bBuf[i] = '\0';
 578 |                 }
 579 |         }
 580 |         a = atoi(aBuf);
 581 |         realA = aTag == 1 ? atoi(realABuf) : -1;
 582 |         b = atoi(bBuf);
 583 |         realB = bTag == 1 ? atoi(realBBuf) : -1;
 584 |         c = atoi(cBuf);
 585 |         d = atoi(dBuf);
 586 | }
 587 | 
 588 | void delta2psl(string st0)
 589 | {
 590 |         ifstream in;
 591 |         ofstream out;
 592 |         string st1, buf, align;
 593 |         int i, j, sp, b, targetID, sourceID, targetSize, sourceSize, targetStart, targetEnd, sourceStart, sourceEnd, sourceGap, targetGap, realSourceID, realTargetID, dummyID;
 594 |         char fr;
 595 |         vector<Segment> seg;
 596 |         Segment s;
 597 | 
 598 |         in.open(st0.c_str());
 599 | 	st1 = st0.substr(0, st0.find('.', st0.find('.', 0) + 1)).append(".psl");
 600 |         out.open(st1.c_str());
 601 | 
 602 |         if(in.is_open())
 603 |         {
 604 |                 getline(in, buf);
 605 |                 getline(in, buf);
 606 |                 while(in.good())
 607 |                 {
 608 |                         align.clear(); seg.clear(); sourceGap = targetGap = 0;
 609 |                         getline(in, buf);
 610 |                         if(buf[0] == '\0') break;
 611 | 
 612 |                         if(buf[0] == '>')
 613 |                         {
 614 |                                 parseDelta(buf, targetID, sourceID, targetSize, sourceSize, realTargetID, realSourceID);
 615 |                                 getline(in, buf);
 616 |                         }
 617 |                         parseDelta(buf, targetStart, targetEnd, sourceStart, sourceEnd, dummyID, dummyID);
 618 |                         if(sourceStart < sourceEnd)
 619 |                                 fr = '+';
 620 |                         else
 621 |                         {
 622 |                                 fr = '-';
 623 |                                 b = sourceStart;
 624 |                                 sourceStart = sourceEnd;
 625 |                                 sourceEnd = b;
 626 |                         }
 627 |                         getline(in, buf);
 628 | 
 629 |                         //generate alignment string
 630 |                         while(buf[0] != '0')
 631 |                         {
 632 |                                 b = atoi(buf.c_str());
 633 |                                 for(i = 1; i < abs(b); i ++)
 634 |                                         align.push_back('M');
 635 |                                 if(b > 0)
 636 |                                 {
 637 |                                         align.push_back('I');
 638 |                                         targetGap ++;
 639 |                                 }
 640 |                                 else
 641 |                                 {
 642 |                                         align.push_back('D');
 643 |                                         sourceGap ++;
 644 |                                 }
 645 |                                 getline(in, buf);
 646 |                         }
 647 |                         for(i = align.size(); i < (sourceEnd - sourceStart + 1) + targetGap; i ++)
 648 |                                 align.push_back('M');
 649 | 
 650 |                         //calculate seg
 651 |                         for(i = 0, j = 0; i < align.size(); i ++)
 652 |                         {
 653 |                                 if(align[i] != 'I')
 654 |                                         j ++;
 655 |                                 if(i == 0 && align[i] == 'M')
 656 |                                 {
 657 |                                         s.sourceStart = sourceStart - 1 + j;
 658 |                                         s.targetStart = s.size = -1;
 659 |                                         seg.push_back(s);
 660 |                                 }
 661 |                                 if(i - 1 >= 0 && (align[i - 1] == 'I' || align[i - 1] == 'D') && align[i] == 'M')
 662 |                                 {
 663 |                                         s.sourceStart = sourceStart - 1 + j;
 664 |                                         s.targetStart = s.size = -1;
 665 |                                         seg.push_back(s);
 666 |                                 }
 667 |                         }
 668 |                         for(i = 0, j = 0, sp = 0; i < align.size(); i ++)
 669 |                         {
 670 |                                 if(align[i] != 'D')
 671 |                                         j ++;
 672 |                                 if(i == 0 && align[i] == 'M')
 673 |                                 {
 674 |                                         seg[sp ++].targetStart = targetStart - 1 + j;
 675 |                                 }
 676 |                                 if(i - 1 >= 0 && (align[i - 1] == 'I' || align[i - 1] == 'D') && align[i] == 'M')
 677 |                                 {
 678 |                                         seg[sp ++].targetStart = targetStart - 1 + j;
 679 |                                 }
 680 |                         }
 681 |                         for(i = 0, j = 0, sp = 0; i < align.size(); i ++)
 682 |                         {
 683 |                                 if(align[i] == 'M')
 684 |                                         j ++;
 685 |                                 if(i + 1 < align.size() && align[i] == 'M' && (align[i + 1] == 'I' || align[i + 1] == 'D'))
 686 |                                 {
 687 |                                         seg[sp ++].size = j;
 688 |                                         j = 0;
 689 |                                 }
 690 |                                 if(i == align.size() - 1 && align[i] == 'M')
 691 |                                         seg[sp].size = j;
 692 |                         }
 693 | 
 694 |                         //write out one psl line
 695 |                         sourceStart --;
 696 |                         targetStart --;
 697 |                         for(i = 0; i < seg.size(); i ++)
 698 |                         {
 699 |                                 seg[i].sourceStart --;
 700 |                                 seg[i].targetStart --;
 701 |                         }
 702 |                         out << "NA	NA	NA	NA	NA	" << sourceGap << "	NA	" << targetGap << "	" << fr << "	";
 703 |                         if(realSourceID != -1)
 704 |                                 out << sourceID << "." << realSourceID << "	";
 705 |                         else
 706 |                                 out << sourceID << "	";
 707 |                         out << sourceSize << "	" << sourceStart << "	" << sourceEnd << "	";
 708 |                         if(realTargetID != -1)
 709 |                                 out << targetID << "." << realTargetID << "	";
 710 |                         else
 711 |                                 out << targetID << "	";
 712 |                         out << targetSize << "	" << targetStart << "	" << targetEnd << "	NA	";
 713 |                         for(i = 0; i < seg.size(); i ++)
 714 |                                 out << seg[i].size << ",";
 715 |                         out << "	";
 716 |                         for(i = 0; i < seg.size(); i ++)
 717 |                                 out << seg[i].sourceStart << ",";
 718 |                         out << "	";
 719 |                         for(i = 0; i < seg.size(); i ++)
 720 |                                 out << seg[i].targetStart << ",";
 721 |                         out << endl;
 722 |                 }
 723 |         }
 724 |         else
 725 |         {
 726 |                 cout << "CANNOT OPEN FILE!" << endl;
 727 |                 exit(-1);
 728 |         }
 729 | }
 730 | 
 731 | int keepPositions(vector<Seq> & contigs, unsigned int sourceID, vector<Segment> & segs, double threshold)
 732 | //This is only used to complete the combination of BLAT's local alignments
 733 | {
 734 | 	int pp, match;
 735 | 
 736 | 	if(sourceID == -1) return 1;
 737 | 	if(contigs[sourceID].positionSets.size() == 0) return 1;
 738 | 	match = 0;
 739 | 	for(pp = 0; pp < contigs[sourceID].positionSets[contigs[sourceID].positionSets.size() - 1].size(); pp ++)
 740 | 	{
 741 | 		if(contigs[sourceID].positionSets[contigs[sourceID].positionSets.size() - 1][pp].chromosomeID != -1)
 742 | 			match ++;
 743 | 	}
 744 | 	if((double) match / contigs[sourceID].positionSets[contigs[sourceID].positionSets.size() - 1].size() >= threshold)
 745 | 		return 1;
 746 | 	else
 747 | 		return 0;
 748 | }
 749 | 
 750 | void complement(vector<Seq> & contigs, unsigned int sourceID)
 751 | {
 752 |         unsigned int np;
 753 | 
 754 |         for(np = 0; np < contigs[sourceID].nucleotides.size(); np ++)
 755 |                 if(contigs[sourceID].nucleotides[np] == 'A') contigs[sourceID].nucleotides[np] = 'T';
 756 |                 else if(contigs[sourceID].nucleotides[np] == 'C') contigs[sourceID].nucleotides[np] = 'G';
 757 |                 else if(contigs[sourceID].nucleotides[np] == 'G') contigs[sourceID].nucleotides[np] = 'C';
 758 |                 else if(contigs[sourceID].nucleotides[np] == 'T') contigs[sourceID].nucleotides[np] = 'A';
 759 |                 //do nothing for 'N'
 760 | }
 761 | 
 762 | static unsigned int sourceIDBak = -1;
 763 | void updateContig(vector<Seq> & contigs, unsigned int sourceID, unsigned int targetID, vector<Segment> & segs, unsigned int fr, double threshold)
 764 | {
 765 | 	int sp, ssp, np;
 766 | 	vector<Position> positions;
 767 | 	Position p;
 768 | 
 769 | 	if(targetID == -1) return;
 770 | 	p.chromosomeID = p.chromosomeOffset = -1;
 771 | 
 772 | 	if(sourceID != sourceIDBak)
 773 | 	{
 774 | 		if(keepPositions(contigs, sourceIDBak, segs, threshold) == 0)
 775 | 		{
 776 | 			contigs[sourceIDBak].positionSets.erase(contigs[sourceIDBak].positionSets.end() - 1);
 777 | 			contigs[sourceIDBak].frSets.erase(contigs[sourceIDBak].frSets.end() - 1);
 778 | 		}
 779 | 		contigs[sourceID].positionSets.push_back(positions);
 780 | 		contigs[sourceID].frSets.push_back(fr);
 781 | 		for(np = 0; np < contigs[sourceID].nucleotides.size(); np ++)
 782 | 			contigs[sourceID].positionSets[contigs[sourceID].positionSets.size() - 1].push_back(p);
 783 | 		//contigs[sourceID].positionSets[contigs[sourceID].positionSets.size() - 1].resize(contigs[sourceID].nucleotides.size(), p);
 784 | 		sourceIDBak = sourceID;
 785 | 	}
 786 | 	else
 787 | 	{
 788 | 		for(sp = 0; sp < segs.size(); sp ++)
 789 | 			for(ssp = segs[sp].sourceStart; ssp < segs[sp].sourceStart + segs[sp].size; ssp ++)
 790 | 			{
 791 | 				if(contigs[sourceID].positionSets[contigs[sourceID].positionSets.size() - 1][ssp].chromosomeID != -1)
 792 | 				{
 793 | 					if(keepPositions(contigs, sourceID, segs, threshold) == 0)
 794 | 					{
 795 | 						contigs[sourceID].positionSets.erase(contigs[sourceID].positionSets.end() - 1);
 796 | 						contigs[sourceID].frSets.erase(contigs[sourceID].frSets.end() - 1);
 797 | 					}
 798 | 					contigs[sourceID].positionSets.push_back(positions);
 799 | 					contigs[sourceID].frSets.push_back(fr);
 800 | 					for(np = 0; np < contigs[sourceID].nucleotides.size(); np ++)
 801 | 						contigs[sourceID].positionSets[contigs[sourceID].positionSets.size() - 1].push_back(p);
 802 | 					//contigs[sourceID].positionSets[contigs[sourceID].positionSets.size() - 1].resize(contigs[sourceID].nucleotides.size(), p);
 803 | 					goto cont;
 804 | 				}
 805 | 			}
 806 | 	}
 807 | 
 808 | cont:
 809 | 	for(sp = 0; sp < segs.size(); sp ++)
 810 | 		for(ssp = 0; ssp < segs[sp].size; ssp ++)
 811 | 		{
 812 | 			contigs[sourceID].positionSets[contigs[sourceID].positionSets.size() - 1][segs[sp].sourceStart + ssp].chromosomeID = targetID;
 813 | 			contigs[sourceID].positionSets[contigs[sourceID].positionSets.size() - 1][segs[sp].sourceStart + ssp].chromosomeOffset = segs[sp].targetStart + ssp;
 814 | 		}		
 815 | }
 816 | 
 817 | void loadContiAli(ifstream & ca, vector<Seq> & contigs, int chromosomeID)
 818 | {
 819 | 	string buf;
 820 | 	unsigned int targetID, targetStart, targetEnd, targetGap, sourceID, sourceStart, sourceEnd, sourceGap, sourceSize, fr, targetSize;
 821 | 	int i, cp, pp, ppp, seqID, realSourceID;
 822 | 	vector<Segment> segs;
 823 | 
 824 | 	sourceID = -1;
 825 | 	if(ca.is_open())
 826 | 	{
 827 | 		while(ca.good())
 828 | 		{
 829 | 			getline(ca, buf);
 830 | 			if(buf[0] == 0) 
 831 | 			{
 832 | 				if(keepPositions(contigs, sourceID, segs, CONTIG_THRESHOLD) == 0)
 833 | 				{
 834 | 					contigs[sourceID].positionSets.erase(contigs[sourceID].positionSets.end() - 1);// have to keep or discard the last alignment
 835 | 				}
 836 | 				break;
 837 | 			}
 838 | 
 839 | 			realSourceID = parseBLAT(buf, targetID, targetStart, targetEnd, targetGap, sourceID, sourceStart, sourceEnd, sourceGap, sourceSize, segs, fr, targetSize);
 840 | 
 841 | 			if((double) (sourceEnd - sourceStart - sourceGap) / sourceSize >= INIT_CONTIG_THRESHOLD && (double) (targetEnd - targetStart - targetGap) / (targetEnd - targetStart) >= INIT_CONTIG_THRESHOLD && sourceSize > 200)
 842 | 			{
 843 | 				updateContig(contigs, sourceID, targetID, segs, fr, CONTIG_THRESHOLD);
 844 | 			}
 845 | 		}
 846 | 	}
 847 | 	else
 848 | 	{
 849 | 		cout << "CANNOT OPEN FILE!" << endl;
 850 | 		exit(-1);
 851 | 	}
 852 | }
 853 | 
 854 | void reverseComplement(vector<char> & seq)
 855 | {
 856 | 	int np;
 857 | 
 858 | 	reverse(seq.begin(), seq.end());
 859 | 	for(np = 0; np < seq.size(); np ++)
 860 | 		if(seq[np] == 'A') seq[np] = 'T';
 861 | 		else if(seq[np] == 'C') seq[np] = 'G';
 862 | 		else if(seq[np] == 'G') seq[np] = 'C';
 863 | 		else if(seq[np] == 'T') seq[np] = 'A';
 864 | 		//do nothing for 'N'
 865 | }
 866 | 
 867 | void printInitialContigs(ofstream & out, Seq & seq, int seqID)
 868 | {
 869 | 	int np;
 870 | 
 871 | 	if(seq.outputted == 0)
 872 | 	{
 873 | 		out << ">" << seqID << endl;
 874 | 		for(np = 0; np < seq.nucleotides.size(); np ++)
 875 | 		{
 876 | 			out << seq.nucleotides[np];
 877 | 			if((np + 1) % 60 ==0 || np == seq.nucleotides.size() - 1)
 878 | 				out << endl;
 879 | 		}
 880 | 		seq.outputted = 1;
 881 | 	}
 882 | }
 883 | 
 884 | void updateGenomeWithContig(vector<Seq> & seqs, vector<vector<Base> > & genome, int chrID)
 885 | {
 886 | 	int sp, pp, ppp, npp, gp, tag, seqID, size, np;
 887 | 	unsigned int chromosomeID, chromosomeOffset, nextID, nextOffset, nextItem, cpp, i, previousIDBak, previousOffsetBak, previousItemBak;
 888 | 	Base b;
 889 | 	ContiMer p, q;
 890 | 	KMer k;
 891 | 	char nucleotide;
 892 | 
 893 | 	previousIDBak = previousOffsetBak = previousItemBak = -1;
 894 | 	for(sp = 0; sp < seqs.size(); sp ++)
 895 | 	{
 896 | 		pp = 0;
 897 | cont:
 898 | 		size = seqs[sp].positionSets.size() == 0 ? 0 : 1;
 899 | 		for(; pp < seqs[sp].positionSets.size() ; pp ++)
 900 | 		{
 901 | 			//for(spp = 0; spp < sp; spp ++) // it is too time-consuming to iterate different seqs with different positions for one seq, so function compatible enables merging kmers from with different contig ids
 902 | 			for(ppp = 0; ppp < pp; ppp ++)
 903 | 				if(abs((int)(seqs[sp].positionSets[pp][0].chromosomeOffset - seqs[sp].positionSets[ppp][0].chromosomeOffset)) < seqs[sp].nucleotides.size())
 904 | 				{
 905 | 					pp ++;
 906 | 					goto cont;
 907 | 				}
 908 | 			for(ppp = 0; ppp < seqs[sp].positionSets[pp].size() - 1; ppp ++)
 909 | 			{
 910 | 				if(seqs[sp].positionSets[pp][ppp].chromosomeID != -1)
 911 | 				{
 912 | 					chromosomeID = seqs[sp].positionSets[pp][ppp].chromosomeID;
 913 | 					chromosomeOffset = seqs[sp].positionSets[pp][ppp].chromosomeOffset;
 914 | 					if(genome[chromosomeID][chromosomeOffset].contiMer.size() >= 2)
 915 | 					{
 916 | 						pp ++;
 917 | 						goto cont;
 918 | 					}
 919 | 				}
 920 | 			}
 921 | 			tag = 0;
 922 | 			if(seqs[sp].frSets[pp] == 1)
 923 | 			{
 924 | 				reverseComplement(seqs[sp].nucleotides);
 925 | 				tag = 1;
 926 | 			}
 927 | //has to always adjust nucleotides according to fr
 928 | 			seqs[sp].outputted = 1;
 929 | //			printInitialContigs(out, seqs[sp], sp);
 930 | 			for(ppp = 0; ppp < seqs[sp].positionSets[pp].size() - 1; ppp ++)
 931 | 			{
 932 | 				if(seqs[sp].positionSets[pp][ppp].chromosomeID != -1)
 933 | 				{
 934 | 					chromosomeID = seqs[sp].positionSets[pp][ppp].chromosomeID;
 935 | 					chromosomeOffset = seqs[sp].positionSets[pp][ppp].chromosomeOffset;
 936 | 					nextID = seqs[sp].positionSets[pp][ppp + 1].chromosomeID;
 937 | 					nextOffset = seqs[sp].positionSets[pp][ppp + 1].chromosomeOffset;
 938 | //					nextItem = genome[nextID][nextOffset].nextPosition.size();
 939 | 					nucleotide = seqs[sp].nucleotides[ppp];
 940 | 					if(nextID == -1)// insertion to genome
 941 | 					{
 942 | 						for(npp = ppp + 2; npp < seqs[sp].positionSets[pp].size(); npp ++)
 943 | 						{
 944 | 							if(seqs[sp].positionSets[pp][npp].chromosomeID != -1)
 945 | 							{
 946 | 								nextID = seqs[sp].positionSets[pp][npp].chromosomeID;
 947 | 								nextOffset = seqs[sp].positionSets[pp][npp].chromosomeOffset;
 948 | 								if(seqs[sp].positionSets[pp][npp].chromosomeID == seqs[sp].positionSets[pp][ppp].chromosomeID && npp - ppp < SI)// small insertion
 949 | 								{
 950 | //									cout << "SI" << endl;
 951 | //									for(cpp = chromosomeOffset; cpp < nextOffset; cpp ++)
 952 | 									{
 953 | 										//p.previousID = previousIDBak;
 954 | 										//p.previousOffset = previousOffsetBak;
 955 | 										//p.previousItem = previousItemBak;
 956 | 										p.nextID = nextID;
 957 | 										p.nextOffset = nextOffset;
 958 | 										p.nextItem = genome[p.nextID][p.nextOffset].contiMer.size();
 959 | 										p.contigID = sp;
 960 | 										p.contigOffset = ppp;
 961 | 										p.nucleotide = nucleotide;
 962 | 										genome[chromosomeID][chromosomeOffset].contiMer.push_back(p);
 963 | 
 964 | 										previousIDBak = chromosomeID;
 965 | 										previousOffsetBak = chromosomeOffset;
 966 | 										previousItemBak = genome[chromosomeID][chromosomeOffset].contiMer.size() - 1;
 967 | 										//k.contigOffset = k.contigOffset0 = k.chromosomeID0 = k.chromosomeOffset0 = -1;
 968 | 										//genome[chromosomeID][chromosomeOffset].kMer.push_back(k);
 969 | 										//genome[chromosomeID][cpp].nextID = chromosomeID;
 970 | 										//genome[chromosomeID][cpp].nextOffset = cpp + 1;
 971 | 									}
 972 | //									cout << "IS" << endl;
 973 | 								}
 974 | 								else// large insertion
 975 | 								{
 976 | //									cout << "LI" << endl;
 977 | 									//p.previousID = previousIDBak;
 978 | 									//p.previousOffset = previousOffsetBak;
 979 | 									//p.previousItem = previousItemBak;
 980 | 									p.nextID = chromosomeID;
 981 | 									p.nextOffset = genome[chromosomeID].size();
 982 | 									p.nextItem = 0;
 983 | 									p.contigID = sp;
 984 | 									p.contigOffset = ppp;
 985 | 									p.nucleotide = nucleotide;
 986 | 									genome[chromosomeID][chromosomeOffset].contiMer.push_back(p);
 987 | 									previousIDBak = chromosomeID;
 988 | 									previousOffsetBak = chromosomeOffset;
 989 | 									previousItemBak = genome[chromosomeID][chromosomeOffset].contiMer.size() - 1;
 990 | 									//k.contigOffset = k.contigOffset0 = k.chromosomeID0 = k.chromosomeOffset0 = -1;
 991 | 									//genome[chromosomeID][chromosomeOffset].kMer.push_back(k);
 992 | 									//genome[chromosomeID][chromosomeOffset].nextID = chromosomeID;
 993 | 									//genome[chromosomeID][chromosomeOffset].nextOffset = genome[chromosomeID].size();
 994 | 									for(i = 0; i < npp - ppp - 2; i ++)
 995 | 									{
 996 | 										b.contiMer.clear();
 997 | 										b.nucleotide = seqs[sp].nucleotides[ppp + 1 + i];
 998 | 										//p.previousID = previousIDBak;
 999 | 										//p.previousOffset = previousOffsetBak;
1000 | 										//p.previousItem = previousItemBak;
1001 | 										p.nextID = chromosomeID;
1002 | 										p.nextOffset = genome[chromosomeID].size() + 1;
1003 | 										p.nextItem = 0;
1004 | 										p.contigID = sp;
1005 | 										p.contigOffset = ppp + 1 + i;
1006 | 										p.nucleotide = b.nucleotide;
1007 | 										b.contiMer.push_back(p);
1008 | 
1009 | 										previousIDBak = chromosomeID;
1010 | 										previousOffsetBak = genome[chromosomeID].size();
1011 | 										previousItemBak = 0;
1012 | 										//k.contigOffset = k.contigOffset0 = k.chromosomeID0 = k.chromosomeOffset0 = -1;
1013 | 										//b.kMer.push_back(k);
1014 | 										//b.nextID = chromosomeID;
1015 | 										//b.nextOffset = genome[chromosomeID].size();
1016 | 										genome[chromosomeID].push_back(b);
1017 | 									}
1018 | 									b.contiMer.clear();
1019 | 									b.nucleotide = seqs[sp].nucleotides[npp - 1];
1020 | 									//p.previousID = previousIDBak;
1021 | 									//p.previousOffset = previousOffsetBak;
1022 | 									//p.previousItem = previousItemBak;
1023 | 									p.nextID = nextID; //contigs[cp].positionSets[pp][npp].chromosomeID;
1024 | 									p.nextOffset = nextOffset; //contigs[cp].positionSets[pp][npp].chromosomeOffset;
1025 | 									p.nextItem = genome[p.nextID][p.nextOffset].contiMer.size();
1026 | 									p.contigID = sp;
1027 | 									p.contigOffset = npp - 1;
1028 | 									p.nucleotide = b.nucleotide;
1029 | 									b.contiMer.push_back(p);
1030 | 
1031 | 									previousIDBak = chromosomeID;
1032 | 									previousOffsetBak = genome[chromosomeID].size();
1033 | 									previousItemBak = 0;
1034 | 									//k.contigOffset = k.contigOffset0 = k.chromosomeID0 = k.chromosomeOffset0 = -1;
1035 | 									//b.kMer.push_back(k);
1036 | 									genome[chromosomeID].push_back(b);
1037 | 									//genome[chromosomeID][genome[chromosomeID].size() - 1].nextID = nextID;
1038 | 									//genome[chromosomeID][genome[chromosomeID].size() - 1].nextOffset = nextOffset;
1039 | //									cout << "IL" << endl;
1040 | 								}
1041 | 								ppp = npp - 1;
1042 | 								break;
1043 | 							}
1044 | 						}
1045 | 					}
1046 | 					else if(nextID == chromosomeID && nextOffset != chromosomeOffset + 1)// deletion from genome
1047 | 					{
1048 | 						if(nextOffset - chromosomeOffset < SD)// small deletion
1049 | 						{
1050 | //							cout << "SD" << endl;
1051 | 							for(cpp = chromosomeOffset; cpp < nextOffset; cpp ++)
1052 | 							{
1053 | 								//p.previousID = previousIDBak;
1054 | 								//p.previousOffset = previousOffsetBak;
1055 | 								//p.previousItem = previousItemBak;
1056 | 								p.nextID = chromosomeID;
1057 | 								p.nextOffset = cpp + 1;
1058 | 								p.nextItem = genome[p.nextID][p.nextOffset].contiMer.size();
1059 | 								p.contigID = sp;
1060 | 								p.contigOffset = ppp;
1061 | 								if(cpp == chromosomeOffset) p.nucleotide = nucleotide;
1062 | 								else p.nucleotide = genome[chromosomeID][cpp].nucleotide;
1063 | 								genome[chromosomeID][cpp].contiMer.push_back(p);
1064 | 
1065 | 								previousIDBak = chromosomeID;
1066 | 								previousOffsetBak = cpp;
1067 | 								previousItemBak = genome[chromosomeID][cpp].contiMer.size() - 1;
1068 | 								//k.contigOffset = k.contigOffset0 = k.chromosomeID0 = k.chromosomeOffset0 = -1;
1069 | 								//genome[chromosomeID][cpp].kMer.push_back(k);
1070 | 								//genome[chromosomeID][cpp].nextID = chromosomeID;
1071 | 								//genome[chromosomeID][cpp].nextOffset = cpp + 1;
1072 | 							}
1073 | //							cout << "DS" << endl;
1074 | 						}
1075 | 						else// large deletion
1076 | 						{
1077 | //							cout << "LD" << endl;
1078 | 							//p.previousID = previousIDBak;
1079 | 							//p.previousOffset = previousOffsetBak;
1080 | 							//p.previousItem = previousItemBak;
1081 | 							p.nextID = nextID;
1082 | 							p.nextOffset = nextOffset;
1083 | 							p.nextItem = genome[p.nextID][p.nextOffset].contiMer.size();
1084 | 							p.contigID = sp;
1085 | 							p.contigOffset = ppp;
1086 | 							p.nucleotide = nucleotide;
1087 | 							genome[chromosomeID][chromosomeOffset].contiMer.push_back(p);
1088 | 
1089 | 							previousIDBak = chromosomeID;
1090 | 							previousOffsetBak = chromosomeOffset;
1091 | 							previousItemBak = genome[chromosomeID][chromosomeOffset].contiMer.size() - 1;
1092 | 							//k.contigOffset = k.contigOffset0 = k.chromosomeID0 = k.chromosomeOffset0 = -1;
1093 | 							//genome[chromosomeID][chromosomeOffset].kMer.push_back(k);
1094 | 							//genome[chromosomeID][chromosomeOffset].nextID = nextID;
1095 | 							//genome[chromosomeID][chromosomeOffset].nextOffset = nextOffset;
1096 | //							cout << "DL" << endl;
1097 | 						}
1098 | 					}
1099 | 					else// ordinary case
1100 | 					{
1101 | 						//p.previousID = previousIDBak;
1102 | 						//p.previousOffset = previousOffsetBak;
1103 | 						//p.previousItem = previousItemBak;
1104 | 						p.nextID = nextID;
1105 | 						p.nextOffset = nextOffset;
1106 | 						p.nextItem = genome[p.nextID][p.nextOffset].contiMer.size();
1107 | 						p.contigID = sp;
1108 | 						p.contigOffset = ppp;
1109 | 						p.nucleotide = nucleotide;
1110 | 						genome[chromosomeID][chromosomeOffset].contiMer.push_back(p);
1111 | 						previousIDBak = chromosomeID;
1112 | 						previousOffsetBak = chromosomeOffset;
1113 | 						previousItemBak = genome[chromosomeID][chromosomeOffset].contiMer.size() - 1;
1114 | 						//k.contigOffset = k.contigOffset0 = k.chromosomeID0 = k.chromosomeOffset0 = -1;
1115 | 						//genome[chromosomeID][chromosomeOffset].kMer.push_back(k);
1116 | 						//genome[chromosomeID][chromosomeOffset].nextID = nextID;
1117 | 						//genome[chromosomeID][chromosomeOffset].nextOffset = nextOffset;
1118 | 					}
1119 | 				}
1120 | 			}
1121 | 			if(nextID != -1)
1122 | 			{
1123 | 				//p.previousID = previousIDBak;
1124 | 				//p.previousOffset = previousOffsetBak;
1125 | 				//p.previousItem = previousItemBak;
1126 | 				p.nextID = p.nextOffset = p.nextItem = -1;
1127 | 				p.contigID = sp;
1128 | 				p.contigOffset = ppp;
1129 | 				p.nucleotide = genome[nextID][nextOffset].nucleotide;
1130 | 				genome[nextID][nextOffset].contiMer.push_back(p);
1131 | 
1132 | 				//k.contigOffset = k.contigOffset0 = k.chromosomeID0 = k.chromosomeOffset0 = -1;
1133 | 				//genome[nextID][nextOffset].kMer.push_back(k);
1134 | 			}
1135 | 			else// chromosomeID != -1 and nextID == -1
1136 | 			{
1137 | 				//p.previousID = previousIDBak;
1138 | 				//p.previousOffset = previousOffsetBak;
1139 | 				//p.previousItem = previousItemBak;
1140 | 				p.nextID = p.nextOffset = p.nextItem = -1;
1141 | 				p.contigID = sp;
1142 | 				p.contigOffset = ppp;
1143 | 				p.nucleotide = genome[chromosomeID][chromosomeOffset].nucleotide;
1144 | 				genome[chromosomeID][chromosomeOffset].contiMer.push_back(p);
1145 | 
1146 | 				//k.contigOffset = k.contigOffset0 = k.chromosomeID0 = k.chromosomeOffset0 = -1;
1147 | 				//genome[chromosomeID][chromosomeOffset].kMer.push_back(k);
1148 | 			}
1149 | 			if(tag == 1) reverseComplement(seqs[sp].nucleotides);
1150 | 		}
1151 | 
1152 | //		cout << "*************************" << endl;
1153 | //		for(gp = 0; gp < genome.size(); gp ++)
1154 | //		{
1155 | //			for(cpp = 0; cpp < genome[gp].size(); cpp ++)
1156 | //			{
1157 | //				cout << "[";
1158 | //				for(pp = 0; pp < genome[gp][cpp].contiMer.size(); pp ++)
1159 | //					cout << "<" << genome[gp][cpp].contiMer[pp].contigOffset << "| " << genome[gp][cpp].contiMer[pp].previousID << ", " << genome[gp][cpp].contiMer[pp].previousOffset << ", " << genome[gp][cpp].contiMer[pp].previousItem << "| " << genome[gp][cpp].contiMer[pp].nextID << ", " << genome[gp][cpp].contiMer[pp].nextOffset << ", " << genome[gp][cpp].contiMer[pp].nextItem << ">";
1160 | //				cout << "]";
1161 | //			}
1162 | //			cout << endl;
1163 | //		}
1164 | //        	cout << "-------------------------" << endl;
1165 | //        	for(gp = 0; gp < genome.size(); gp ++)
1166 | //        	{
1167 | //                	for(cpp = 0; cpp < genome[gp].size(); cpp ++)
1168 | //                	{
1169 | //                        	cout << "[";
1170 | //                        	for(pp = 0; pp < genome[gp][cpp].kMer.size(); pp ++)
1171 | //                                	cout << "<" << genome[gp][cpp].kMer[pp].contigOffset << ", " << genome[gp][cpp].kMer[pp].contigOffset0 << ", " << genome[gp][cpp].kMer[pp].chromosomeID0 << ", " << genome[gp][cpp].kMer[pp].chromosomeOffset0 <<  "| " << genome[gp][cpp].kMer[pp].previous[0].previousID << ", " << genome[gp][cpp].kMer[pp].previous[0].previousOffset << ", " << genome[gp][cpp].kMer[pp].previous[0].previousItem << "| " << genome[gp][cpp].kMer[pp].next[0].nextID << ", " << genome[gp][cpp].kMer[pp].next[0].nextOffset << ", " << genome[gp][cpp].kMer[pp].next[0].nextItem << ">";
1172 | //                        	cout << "]";
1173 | //                	}
1174 | //        	}
1175 | //		cout << endl;
1176 | //		cout << "*************************" << endl;
1177 | 	}
1178 | 
1179 | 	ofstream out;
1180 |         string s = "tmp/_initial_contigs." + itoa(chrID) + ".fa";
1181 |         out.open(s.c_str());
1182 | 	vector<vector<char> > contigs;
1183 | 	vector<char> contig;
1184 | 	vector<int> outputted;
1185 | 	vector<int> maxOutputted;
1186 | 	int cp, frag;
1187 | 
1188 | 	int IDBak = -1;
1189 |         for(sp = 0; sp < seqs.size(); sp ++)
1190 |         {
1191 | 		if(seqs[sp].ID != IDBak)
1192 | 		{
1193 | 			maxOutputted.push_back(0);
1194 | 			outputted.push_back(0);
1195 | 			contigs.push_back(contig);
1196 | 			IDBak = seqs[sp].ID;
1197 | 		}
1198 | 		maxOutputted[maxOutputted.size() - 1] ++;
1199 | 		outputted[outputted.size() - 1] = outputted[outputted.size() - 1] + seqs[sp].outputted;
1200 | 		for(np = 0; np < seqs[sp].nucleotides.size(); np ++)
1201 | 			contigs[contigs.size() - 1].push_back(seqs[sp].nucleotides[np]);
1202 | 	}
1203 | 
1204 | 	for(cp = 0; cp < contigs.size(); cp ++)
1205 | 	{
1206 | 		if((double) outputted[cp] / (double) maxOutputted[cp] >= CONTIG_THRESHOLD)
1207 | 		{
1208 | 	                out << ">" << cp << endl;
1209 | 	                for(np = 0; np < contigs[cp].size(); np ++)
1210 | 	                {
1211 | 	                        out << contigs[cp][np];
1212 | 	                        if((np + 1) % 60 == 0 || np == contigs[cp].size() - 1)
1213 | 	                                out << endl;
1214 | 	                }
1215 | 		}
1216 |         }
1217 | }
1218 | 
1219 | void loadContigAlignment(vector<vector<Base> > & genome, int chromosomeID)
1220 | {
1221 |         vector<Seq> contigs;
1222 | 	ifstream c, ca;
1223 | 	string s;
1224 | 
1225 | 	c.open("tmp/_contigs.fa");
1226 | 	s = "tmp/_contigs_genome." + itoa(chromosomeID) + ".psl";
1227 | 	ca.open(s.c_str());
1228 |         loadSeq(c, contigs);
1229 |         loadContiAli(ca, contigs, chromosomeID);
1230 |         updateGenomeWithContig(contigs, genome, chromosomeID);
1231 | }
1232 | 
1233 | int loadReadAli(ifstream & ra, vector<Seq> & reads, int & aliStartID, int & seqStartID)
1234 | {
1235 | 	string buf;
1236 | 	unsigned int targetID1, targetStart1, targetEnd1, targetGap1, sourceID1, sourceStart1, sourceEnd1, sourceGap1, sourceSize1, fr1, targetID2, targetStart2, targetEnd2, targetGap2, sourceID2, sourceStart2, sourceEnd2, sourceGap2, sourceSize2, fr2;
1237 | 	vector<Segment> segs1, segs2;
1238 | 	int gp, cpp, pp;
1239 | 	int count = 0;
1240 | 
1241 | 	sourceIDBak = -1;
1242 | 	if(ra.is_open())
1243 | 	{
1244 | 		while(ra.good())
1245 | 		{
1246 | 			getline(ra, buf);
1247 | 			if(buf[0] == 0) break;
1248 | 			if(buf[0] == '@') continue;
1249 | 			parseBOWTIE(buf, targetID1, targetStart1, targetEnd1, targetGap1, sourceID1, sourceStart1, sourceEnd1, sourceGap1, sourceSize1, segs1, fr1);
1250 | 			getline(ra, buf);
1251 | 			if(buf[0] == 0)
1252 | 			{
1253 | 				cout << "BROKEN BOWTIE FILE" << endl;
1254 | 				exit(-1);
1255 | 			}
1256 | 			parseBOWTIE(buf, targetID2, targetStart2, targetEnd2, targetGap2, sourceID2, sourceStart2, sourceEnd2, sourceGap2, sourceSize2, segs2, fr2);
1257 | 
1258 | 			if(sourceID1 < aliStartID) continue;
1259 | 			if(sourceID1 > seqStartID) return 0;
1260 | 
1261 | 			if(targetID1 != -1 && targetID2 != -1 && (double) (sourceEnd1 - sourceStart1 - sourceGap1) / /*(sourceEnd1 - sourceStart1)*/ sourceSize1 >= THRESHOLD && (double) (targetEnd1 - targetStart1 - targetGap1) / (targetEnd1 - targetStart1) >= THRESHOLD && (double) (sourceEnd2 - sourceStart2 - sourceGap2) / /*(sourceEnd2 - sourceStart2)*/ sourceSize2 >= THRESHOLD && (double) (targetEnd2 - targetStart2 - targetGap2) / (targetEnd2 - targetStart2) >= THRESHOLD)
1262 | 			{
1263 | 				updateContig(reads, (sourceID1 - aliStartID) * 2, targetID1, segs1, fr1, THRESHOLD);
1264 | 				updateContig(reads, (sourceID2 - aliStartID) * 2 + 1, targetID2, segs2, fr2, THRESHOLD);
1265 | 				count ++;
1266 | 			}
1267 | 			segs1.clear();
1268 | 			segs2.clear();
1269 | 		}
1270 | 		return 1;
1271 | 	}
1272 | 	else
1273 | 	{
1274 | 		cout << "CANNOT OPEN FILE!" << endl;
1275 | 		exit(-1);
1276 | 	}
1277 | }
1278 | 
1279 | int getDiff(vector<char> s1, vector<char> s2)
1280 | {
1281 | 	int diff = 0, i;
1282 | 
1283 | 	if(s1.size() == 0 || s2.size() == 0) return -1;
1284 | 
1285 | 	for(i = 0; i < s1.size(); i ++)
1286 | 		if(s1[i] != s2[i])
1287 | 			diff ++;
1288 | 	return diff;
1289 | }
1290 | 
1291 | //allowed pair distance < distanceHigh - distanceLow so that PE de Bruijn graph's branch solving function works
1292 | //distanceHigh - distanceLow > 50
1293 | int compatible(KMer & k1, KMer & k2, int insertVariation, int mrl)
1294 | {
1295 | 	if(
1296 | 	(k1.contigID == -1 || k2.contigID == -1 || (k1.contigID != -1 && k1.contigID == k2.contigID && abs((int)(k1.contigOffset - k2.contigOffset)) <= 5 * EP) 
1297 | #ifdef OPTIMIZATION
1298 | 	|| (k1.contigID != -1 && k2.contigID != -1 && k1.contigID != k2.contigID)
1299 | #endif
1300 | 	) &&
1301 | // it is hard to judge if two overlapping contigs should be joined or are just repetitive. Here I find it performs better to join them.
1302 | 	(k1.contigID0 == -1 || k2.contigID0 == -1 || (k1.contigID0 != -1 && k1.contigID0 == k2.contigID0 && abs((int)(k1.contigOffset0 - k2.contigOffset0)) <= 2 * insertVariation + 5 * EP) 
1303 | #ifdef OPTIMIZATION
1304 | 	|| (k1.contigID0 != -1 && k2.contigID0 != -1 && k1.contigID0 != k2.contigID0)
1305 | #endif
1306 | 	) &&
1307 | 	(k1.chromosomeID0 == -1 || k2.chromosomeID0 == -1 || (k1.chromosomeID0 != -1 && k1.chromosomeID0 == k2.chromosomeID0 && abs((int)(k1.chromosomeOffset0 - k2.chromosomeOffset0)) <= 2 * insertVariation + 5 * EP))
1308 | 	)
1309 | 		return 1;
1310 | 	else
1311 | 		return 0;
1312 | }
1313 | 
1314 | int nextCompatible(Next next1, Next next2)
1315 | {
1316 | 	if(next1.nextID == -1 || next2.nextID == -1)
1317 | 	{
1318 | 		cout << "KMER ERROR" << endl;
1319 | 		exit(-1);
1320 | 	}
1321 | 
1322 | 	if(next1.nextID == next2.nextID && next1.nextOffset == next2.nextOffset && next1.nextItem == next2.nextItem)
1323 | 		return 1;
1324 | 	return 0;
1325 | }
1326 | 
1327 | int previousCompatible(Previous previous1, Previous previous2)
1328 | {
1329 |         if(previous1.previousID == -1 || previous2.previousID == -1)
1330 |         {
1331 |                 cout << "KMER ERROR" << endl;
1332 |                 exit(-1);
1333 |         }
1334 | 
1335 |         if(previous1.previousID == previous2.previousID && previous1.previousOffset == previous2.previousOffset && previous1.previousItem == previous2.previousItem)
1336 |                 return 1;
1337 |         return 0;
1338 | }
1339 | 
1340 | void updateKBases(KMer k, KMer & k0)
1341 | {
1342 | 	if(k.s.size() > 0)
1343 | 	        switch(k.s[0])
1344 | 	        {
1345 | 	                case 'A': k0.A ++; break;
1346 | 	                case 'C': k0.C ++; break;
1347 | 	                case 'G': k0.G ++; break;
1348 | 	                case 'T': k0.T ++; break;
1349 | 	                default: k0.N ++;
1350 | 	        }
1351 | }
1352 | 
1353 | void updateKMer(vector<vector<Base> > & genome, unsigned int chromosomeID, unsigned int chromosomeOffset, unsigned int nextID, unsigned int nextOffset, unsigned int chromosomeID0, unsigned int chromosomeOffset0, unsigned int nextID0, unsigned int nextOffset0, vector<char> s, vector<char> nextS, int insertVariation, int mrl)
1354 | {
1355 | 	int ip, ipp, ip0, np, pp, cip, nip, size;
1356 | 	KMer k1, k2;
1357 | 	ContiMer p;
1358 | 	vector<unsigned int> nextItem, chromosomeItem;
1359 | 	Next next;
1360 | 	Previous previous;
1361 | 
1362 | 	k1.traversed = 0;
1363 | 	k1.s = s;
1364 | 	k1.chromosomeID0 = chromosomeID0;
1365 | 	k1.chromosomeOffset0 = chromosomeOffset0;
1366 | 	k1.A = k1.C = k1.G = k1.T = k1.N = 0;
1367 | 	k1.coverage = 1;
1368 | 
1369 | 	if(genome[chromosomeID][chromosomeOffset].contiMer.size() == 0 && (chromosomeID0 == -1 || genome[chromosomeID0][chromosomeOffset0].contiMer.size() == 0))
1370 | 	{
1371 | 		k1.contigID = -1;
1372 | 		k1.contigID0 = -1;
1373 | 		k1.contigOffset = -1;
1374 | 		k1.contigOffset0 = -1;
1375 | 		for(ipp = 0; ipp < genome[chromosomeID][chromosomeOffset].kMer.size(); ipp ++)
1376 | 			if(compatible(k1, genome[chromosomeID][chromosomeOffset].kMer[ipp], insertVariation, mrl))
1377 | 				break;
1378 | 		if(ipp == genome[chromosomeID][chromosomeOffset].kMer.size())
1379 | 		{
1380 | //			cout << "insert to " << chromosomeID << ", " << chromosomeOffset << ", " << ipp << endl;
1381 | 			genome[chromosomeID][chromosomeOffset].kMer.push_back(k1);
1382 | 			updateKBases(k1, genome[chromosomeID][chromosomeOffset].kMer[ipp]);
1383 | 		}
1384 | 		else
1385 | 		{
1386 | 			genome[chromosomeID][chromosomeOffset].kMer[ipp].coverage ++;
1387 | //			cout << "update " << chromosomeID << ", " << chromosomeOffset << ", " << ipp << endl;
1388 | 			updateKBases(k1, genome[chromosomeID][chromosomeOffset].kMer[ipp]);
1389 | 		}
1390 | 		chromosomeItem.push_back(ipp);
1391 | 		goto cont;
1392 | 	}
1393 | 
1394 | 	if(genome[chromosomeID][chromosomeOffset].contiMer.size() != 0 && (chromosomeID0 == -1 || genome[chromosomeID0][chromosomeOffset0].contiMer.size() == 0))
1395 | 	{
1396 | 		k1.contigID0 = -1;
1397 | 		k1.contigOffset0 = -1;
1398 | 		for(ip = 0; ip < genome[chromosomeID][chromosomeOffset].contiMer.size(); ip ++)
1399 | 		{
1400 | 			k1.contigID = genome[chromosomeID][chromosomeOffset].contiMer[ip].contigID;
1401 | 			k1.contigOffset = genome[chromosomeID][chromosomeOffset].contiMer[ip].contigOffset;
1402 | 			for(ipp = 0; ipp < genome[chromosomeID][chromosomeOffset].kMer.size(); ipp ++)
1403 | 				if(compatible(k1, genome[chromosomeID][chromosomeOffset].kMer[ipp], insertVariation, mrl))
1404 | 					break;
1405 | 			if(ipp == genome[chromosomeID][chromosomeOffset].kMer.size())
1406 | 			{
1407 | //				cout << "insert to " << chromosomeID << ", " << chromosomeOffset << ", " << ipp << endl;
1408 | 				genome[chromosomeID][chromosomeOffset].kMer.push_back(k1);
1409 | 				updateKBases(k1, genome[chromosomeID][chromosomeOffset].kMer[ipp]);
1410 | 			}
1411 | 			else
1412 | 			{
1413 | 				genome[chromosomeID][chromosomeOffset].kMer[ipp].coverage ++;
1414 | //				cout << "update " << chromosomeID << ", " << chromosomeOffset << ", " << ipp << endl;
1415 | 				updateKBases(k1, genome[chromosomeID][chromosomeOffset].kMer[ipp]);
1416 | 			}
1417 | 			chromosomeItem.push_back(ipp);
1418 | 		}
1419 | 		goto cont;
1420 | 	}
1421 | 
1422 | 	if(genome[chromosomeID][chromosomeOffset].contiMer.size() == 0 && (chromosomeID0 != -1 && genome[chromosomeID0][chromosomeOffset0].contiMer.size() != 0))
1423 | 	{
1424 | 		k1.contigID = -1;
1425 | 		k1.contigOffset = -1;
1426 | 		for(ip0 = 0; ip0 < genome[chromosomeID0][chromosomeOffset0].contiMer.size(); ip0 ++)
1427 | 		{
1428 | 			k1.contigID0 = genome[chromosomeID0][chromosomeOffset0].contiMer[ip0].contigID;
1429 | 			k1.contigOffset0 = genome[chromosomeID0][chromosomeOffset0].contiMer[ip0].contigOffset;
1430 | 			for(ipp = 0; ipp < genome[chromosomeID][chromosomeOffset].kMer.size(); ipp ++)
1431 | 				if(compatible(k1, genome[chromosomeID][chromosomeOffset].kMer[ipp], insertVariation, mrl))
1432 | 					break;
1433 | 			if(ipp == genome[chromosomeID][chromosomeOffset].kMer.size())
1434 | 			{
1435 | //				cout << "insert to " << chromosomeID << ", " << chromosomeOffset << ", " << ipp << endl;
1436 | 				genome[chromosomeID][chromosomeOffset].kMer.push_back(k1);
1437 | 				updateKBases(k1, genome[chromosomeID][chromosomeOffset].kMer[ipp]);
1438 | 			}
1439 | 			else
1440 | 			{
1441 | 				genome[chromosomeID][chromosomeOffset].kMer[ipp].coverage ++;
1442 | //				cout << "update " << chromosomeID << ", " << chromosomeOffset << ", " << ipp << endl;
1443 | 				updateKBases(k1, genome[chromosomeID][chromosomeOffset].kMer[ipp]);
1444 | 			}
1445 | 			chromosomeItem.push_back(ipp);
1446 | 		}
1447 | 		goto cont;
1448 | 	}
1449 | 
1450 | 	if(genome[chromosomeID][chromosomeOffset].contiMer.size() != 0 && (chromosomeID0 != -1 && genome[chromosomeID0][chromosomeOffset0].contiMer.size() != 0))
1451 | 	{
1452 | 		for(ip = 0; ip < genome[chromosomeID][chromosomeOffset].contiMer.size(); ip ++)
1453 | 			for(ip0 = 0; ip0 < genome[chromosomeID0][chromosomeOffset0].contiMer.size(); ip0 ++)
1454 | 			{
1455 | 				k1.contigID = genome[chromosomeID][chromosomeOffset].contiMer[ip].contigID;
1456 | 				k1.contigID0 = genome[chromosomeID0][chromosomeOffset0].contiMer[ip0].contigID;
1457 | 				k1.contigOffset = genome[chromosomeID][chromosomeOffset].contiMer[ip].contigOffset;
1458 | 				k1.contigOffset0 = genome[chromosomeID0][chromosomeOffset0].contiMer[ip0].contigOffset;
1459 | 				for(ipp = 0; ipp < genome[chromosomeID][chromosomeOffset].kMer.size(); ipp ++)
1460 | 					if(compatible(k1, genome[chromosomeID][chromosomeOffset].kMer[ipp], insertVariation, mrl))
1461 | 						break;
1462 | 				if(ipp == genome[chromosomeID][chromosomeOffset].kMer.size())
1463 | 				{
1464 | //					cout << "insert to " << chromosomeID << ", " << chromosomeOffset << ", " << ipp << endl;
1465 | 					genome[chromosomeID][chromosomeOffset].kMer.push_back(k1);
1466 | 					updateKBases(k1, genome[chromosomeID][chromosomeOffset].kMer[ipp]);
1467 | 				}
1468 | 				else
1469 | 				{
1470 | 					genome[chromosomeID][chromosomeOffset].kMer[ipp].coverage ++;
1471 | //					cout << "update " << chromosomeID << ", " << chromosomeOffset << ", " << ipp << endl;
1472 | 					updateKBases(k1, genome[chromosomeID][chromosomeOffset].kMer[ipp]);
1473 | 				}
1474 | 				chromosomeItem.push_back(ipp);
1475 | 			}
1476 | 		goto cont;
1477 | 	}
1478 | 
1479 | cont:
1480 | 	k2.traversed = 0;
1481 | 	k2.s = nextS;
1482 |         k2.chromosomeID0 = nextID0;
1483 |         k2.chromosomeOffset0 = nextOffset0;
1484 | 	k2.coverage = 0;
1485 | 	k2.A = k2.C = k2.G = k2.T = k2.N = 0;
1486 | 
1487 |         if(genome[nextID][nextOffset].contiMer.size() == 0 && (nextID0 == -1 || genome[nextID0][nextOffset0].contiMer.size() == 0))
1488 |         {
1489 | 		k2.contigID = -1;
1490 | 		k2.contigID0 = -1;
1491 |                 k2.contigOffset = -1;
1492 |                 k2.contigOffset0 = -1;
1493 |                 for(ipp = 0; ipp < genome[nextID][nextOffset].kMer.size(); ipp ++)
1494 |                         if(compatible(k2, genome[nextID][nextOffset].kMer[ipp], insertVariation, mrl))
1495 |                                 break;
1496 |                 if(ipp == genome[nextID][nextOffset].kMer.size())
1497 |                 {
1498 | //                      cout << "insert to " << nextID << ", " << nextOffset << ", " << ipp << endl;
1499 |                         genome[nextID][nextOffset].kMer.push_back(k2);
1500 |                 }
1501 |                 else
1502 | 		{
1503 | //			genome[chromosomeID][chromosomeOffset].kMer[ipp].coverage ++;
1504 | //                      cout << "update " << nextID << ", " << nextOffset << ", " << ipp << endl;
1505 | 		}
1506 | 		nextItem.push_back(ipp);
1507 | 		goto conti;
1508 |         }
1509 | 
1510 |         if(genome[nextID][nextOffset].contiMer.size() != 0 && (nextID0 == -1 || genome[nextID0][nextOffset0].contiMer.size() == 0))
1511 |         {
1512 | 		k2.contigID0 = -1;
1513 |                 k2.contigOffset0 = -1;
1514 |                 for(ip = 0; ip < genome[nextID][nextOffset].contiMer.size(); ip ++)
1515 |                 {
1516 | 			k2.contigID = genome[nextID][nextOffset].contiMer[ip].contigID;
1517 |                         k2.contigOffset = genome[nextID][nextOffset].contiMer[ip].contigOffset;
1518 |                         for(ipp = 0; ipp < genome[nextID][nextOffset].kMer.size(); ipp ++)
1519 |                                 if(compatible(k2, genome[nextID][nextOffset].kMer[ipp], insertVariation, mrl))
1520 |                                         break;
1521 |                         if(ipp == genome[nextID][nextOffset].kMer.size())
1522 |                         {
1523 | //                              cout << "insert to " << nextID << ", " << nextOffset << ", " << ipp << endl;
1524 |                                 genome[nextID][nextOffset].kMer.push_back(k2);
1525 |                         }
1526 |                         else
1527 | 			{
1528 | //				genome[chromosomeID][chromosomeOffset].kMer[ipp].coverage ++;
1529 | //                              cout << "update " << nextID << ", " << nextOffset << ", " << ipp << endl;
1530 | 			}
1531 | 			nextItem.push_back(ipp);
1532 |                 }
1533 | 		goto conti;
1534 |         }
1535 | 
1536 |         if(genome[nextID][nextOffset].contiMer.size() == 0 && (nextID0 != -1 && genome[nextID0][nextOffset0].contiMer.size() != 0))
1537 |         {
1538 | 		k2.contigID = -1;
1539 |                 k2.contigOffset = -1;
1540 |                 for(ip0 = 0; ip0 < genome[nextID0][nextOffset0].contiMer.size(); ip0 ++)
1541 |                 {
1542 | 			k2.contigID0 = genome[nextID0][nextOffset0].contiMer[ip0].contigID;
1543 |                         k2.contigOffset0 = genome[nextID0][nextOffset0].contiMer[ip0].contigOffset;
1544 |                         for(ipp = 0; ipp < genome[nextID][nextOffset].kMer.size(); ipp ++)
1545 |                                 if(compatible(k2, genome[nextID][nextOffset].kMer[ipp], insertVariation, mrl))
1546 |                                         break;
1547 |                         if(ipp == genome[nextID][nextOffset].kMer.size())
1548 |                         {
1549 | //                              cout << "insert to " << nextID << ", " << nextOffset << ", " << ipp << endl;
1550 |                                 genome[nextID][nextOffset].kMer.push_back(k2);
1551 |                         }
1552 |                         else
1553 | 			{
1554 | //				genome[chromosomeID][chromosomeOffset].kMer[ipp].coverage ++;
1555 | //                              cout << "update " << nextID << ", " << nextOffset << ", " << ipp << endl;
1556 | 			}
1557 | 			nextItem.push_back(ipp);
1558 |                 }
1559 | 		goto conti;
1560 |         }
1561 | 
1562 |         if(genome[nextID][nextOffset].contiMer.size() != 0 && (nextID0 != -1 && genome[nextID0][nextOffset0].contiMer.size() != 0))
1563 | 	{
1564 |                 for(ip = 0; ip < genome[nextID][nextOffset].contiMer.size(); ip ++)
1565 |                         for(ip0 = 0; ip0 < genome[nextID0][nextOffset0].contiMer.size(); ip0 ++)
1566 |                         {
1567 | 				k2.contigID = genome[nextID][nextOffset].contiMer[ip].contigID;
1568 | 				k2.contigID0 = genome[nextID0][nextOffset0].contiMer[ip0].contigID;
1569 |                                 k2.contigOffset = genome[nextID][nextOffset].contiMer[ip].contigOffset;
1570 |                                 k2.contigOffset0 = genome[nextID0][nextOffset0].contiMer[ip0].contigOffset;
1571 |                                 for(ipp = 0; ipp < genome[nextID][nextOffset].kMer.size(); ipp ++)
1572 |                                         if(compatible(k2, genome[nextID][nextOffset].kMer[ipp], insertVariation, mrl))
1573 |                                                 break;
1574 |                                 if(ipp == genome[nextID][nextOffset].kMer.size())
1575 |                                 {
1576 | //                                    cout << "insert to " << nextID << ", " << nextOffset << ", " << ipp << endl;
1577 |                                         genome[nextID][nextOffset].kMer.push_back(k2);
1578 |                                 }
1579 |                                 else
1580 | 				{
1581 | //					genome[chromosomeID][chromosomeOffset].kMer[ipp].coverage ++;
1582 | //                                    cout << "update " << nextID << ", " << nextOffset << ", " << ipp << endl;
1583 | 				}
1584 | 				nextItem.push_back(ipp);
1585 |                         }
1586 | 		goto conti;
1587 | 	}
1588 | //it is important to consider the case that one read geneartes two or more k-mers
1589 | conti:
1590 | 	for(cip = 0; cip < chromosomeItem.size(); cip ++)
1591 | 		for(nip = 0; nip < nextItem.size(); nip ++)
1592 | 		{
1593 | 			next.nextID = nextID;
1594 | 			next.nextOffset = nextOffset;
1595 | 			next.nextItem = nextItem[nip];
1596 | 			for(np = 0; np < genome[chromosomeID][chromosomeOffset].kMer[chromosomeItem[cip]].next.size(); np ++)
1597 | 				if(nextCompatible(next, genome[chromosomeID][chromosomeOffset].kMer[chromosomeItem[cip]].next[np]))
1598 | 					break;
1599 | //decide connectivity between m new k1's and n new k2's
1600 | 			if(np == genome[chromosomeID][chromosomeOffset].kMer[chromosomeItem[cip]].next.size() && 
1601 | 
1602 | 			(genome[nextID][nextOffset].kMer[nextItem[nip]].contigID == -1 || 
1603 | 			genome[chromosomeID][chromosomeOffset].kMer[chromosomeItem[cip]].contigID == -1 || 
1604 | 			(genome[nextID][nextOffset].kMer[nextItem[nip]].contigID != -1 && genome[nextID][nextOffset].kMer[nextItem[nip]].contigID == genome[chromosomeID][chromosomeOffset].kMer[chromosomeItem[cip]].contigID && abs((int)(genome[nextID][nextOffset].kMer[nextItem[nip]].contigOffset - genome[chromosomeID][chromosomeOffset].kMer[chromosomeItem[cip]].contigOffset)) <= 5 * EP) 
1605 | #ifdef OPTIMIZATION
1606 | 			|| genome[nextID][nextOffset].kMer[nextItem[nip]].contigID != -1 && genome[chromosomeID][chromosomeOffset].kMer[chromosomeItem[cip]].contigID != -1 && genome[nextID][nextOffset].kMer[nextItem[nip]].contigID != genome[chromosomeID][chromosomeOffset].kMer[chromosomeItem[cip]].contigID 
1607 | #endif
1608 | 			) &&
1609 |                         (genome[nextID][nextOffset].kMer[nextItem[nip]].contigID0 == -1 ||
1610 |                         genome[chromosomeID][chromosomeOffset].kMer[chromosomeItem[cip]].contigID0 == -1 ||
1611 |                         (genome[nextID][nextOffset].kMer[nextItem[nip]].contigID0 != -1 && genome[nextID][nextOffset].kMer[nextItem[nip]].contigID0 == genome[chromosomeID][chromosomeOffset].kMer[chromosomeItem[cip]].contigID0 && abs((int)(genome[nextID][nextOffset].kMer[nextItem[nip]].contigOffset0 - genome[chromosomeID][chromosomeOffset].kMer[chromosomeItem[cip]].contigOffset0)) <= 2 * insertVariation + 5 * EP) 
1612 | #ifdef OPTIMIZATION
1613 | 			|| genome[nextID][nextOffset].kMer[nextItem[nip]].contigID0 != -1 && genome[chromosomeID][chromosomeOffset].kMer[chromosomeItem[cip]].contigID0 != -1 && genome[nextID][nextOffset].kMer[nextItem[nip]].contigID0 != genome[chromosomeID][chromosomeOffset].kMer[chromosomeItem[cip]].contigID0
1614 | #endif
1615 | 			))
1616 | 
1617 | 			{
1618 | //				cout << "connect [" << chromosomeID << ", " << chromosomeOffset << ", " << cip << ", " << np << "] to [" << nextID << ", " << nextOffset << ", " << nip << "]" << endl;
1619 | 				genome[chromosomeID][chromosomeOffset].kMer[chromosomeItem[cip]].next.push_back(next);
1620 | 			}
1621 | //			else
1622 | //				cout << "keep [" << chromosomeID << ", " << chromosomeOffset << ", " << cip << ", " << np << "] to [" << nextID << ", " << nextOffset << ", " << nip << "]" << endl;
1623 | 		}
1624 | }
1625 | 
1626 | void switchSeqs(vector<Position> & p1, vector<Position> & p2, vector<char> & n1, vector<char> & n2)
1627 | {
1628 | 	vector<Position> tp;
1629 | 	vector<char> tn;
1630 | 
1631 | 	tp = p1; p1 = p2; p2 = tp;
1632 | 	tn = n1; n1 = n2; n2 = tn;
1633 | }
1634 | 
1635 | void updateGenomeWithRead(vector<Seq> & seqs, vector<vector<Base> > & genome, int length, int insertVariation, int mrl)
1636 | {
1637 |         int sp, pp, ppp, npp, gp, sp0, ip, ipp, np, tag0, tag1, tag2;// sp0 is for the other read
1638 |         unsigned int chromosomeID, chromosomeOffset, nextID, nextOffset, nextItem, cpp, i, chromosomeID0, chromosomeOffset0, nextID0, nextOffset0, size;
1639 |         Base b;
1640 |         ContiMer p;
1641 | 	KMer k;
1642 | 	vector<char> s, nextS;
1643 | 
1644 |         for(sp = 0, sp0 = 1; sp < seqs.size(); sp = sp + 2, sp0 = sp0 + 2)
1645 | 	{
1646 | 		pp = 0;
1647 | cont:
1648 |                 for(; pp < seqs[sp].positionSets.size(); pp ++)
1649 |                 {
1650 |                         for(ppp = 0; ppp < pp; ppp ++)
1651 |                                 if(abs((int)(seqs[sp].positionSets[pp][0].chromosomeOffset - seqs[sp].positionSets[ppp][0].chromosomeOffset)) < seqs[sp].nucleotides.size())
1652 | 				{
1653 | 					pp ++;
1654 |                                         goto cont;
1655 | 				}
1656 | 			tag0 = tag1 = tag2 = 0;
1657 |                         if(seqs[sp].frSets[pp] == 1 && seqs[sp0].frSets[pp] == 0)
1658 |                         {
1659 |                                 reverseComplement(seqs[sp].nucleotides);
1660 | 				tag0 = 1;
1661 | 			}
1662 | 			else if(seqs[sp0].frSets[pp] == 1 && seqs[sp].frSets[pp] == 0)
1663 | 			{
1664 | 				reverseComplement(seqs[sp0].nucleotides);
1665 | 				tag1 = 1;
1666 |                         }
1667 | 			else
1668 | 			{
1669 | 				cout << "BOWTIE ALIGNMENT ERROR" << endl;
1670 | 				exit(-1);
1671 | 			}
1672 | 			for(ppp = 0; ppp < seqs[sp].positionSets[pp].size() - length; ppp ++)
1673 | 				if(seqs[sp].positionSets[pp][ppp].chromosomeID != -1 && seqs[sp0].positionSets[pp][ppp].chromosomeID != -1 &&
1674 | 				seqs[sp].positionSets[pp][ppp].chromosomeOffset > seqs[sp0].positionSets[pp][ppp].chromosomeOffset)
1675 | 				{
1676 | 					switchSeqs(seqs[sp].positionSets[pp], seqs[sp0].positionSets[pp], seqs[sp].nucleotides, seqs[sp0].nucleotides);
1677 | 					tag2 = 1;
1678 | 					break;
1679 | 				}
1680 | //has to always adjust nucleotides according to fr to accomodate reverse complements
1681 |                         for(ppp = 0; ppp < seqs[sp].positionSets[pp].size() - length; ppp ++)
1682 |                         {
1683 |                                 if(seqs[sp].positionSets[pp][ppp].chromosomeID != -1)
1684 |                                 {
1685 |                                         chromosomeID = seqs[sp].positionSets[pp][ppp].chromosomeID;
1686 |                                         chromosomeOffset = seqs[sp].positionSets[pp][ppp].chromosomeOffset;
1687 | 					chromosomeID0 = seqs[sp0].positionSets[pp][ppp].chromosomeID;
1688 | 					chromosomeOffset0 = seqs[sp0].positionSets[pp][ppp].chromosomeOffset;
1689 | 
1690 |                                         nextID = seqs[sp].positionSets[pp][ppp + 1].chromosomeID;
1691 |                                         nextOffset = seqs[sp].positionSets[pp][ppp + 1].chromosomeOffset;
1692 | 					nextID0 = seqs[sp0].positionSets[pp][ppp + 1].chromosomeID;
1693 | 					nextOffset0 = seqs[sp0].positionSets[pp][ppp + 1].chromosomeOffset;
1694 | //                                      nextItem = genome[nextID][nextOffset].nextPosition.size();
1695 |                                         if(nextID == -1)// insertion to genome
1696 |                                         {
1697 |                                                 for(npp = ppp + 2; npp < seqs[sp].positionSets[pp].size(); npp ++)
1698 |                                                 {
1699 |                                                         if(seqs[sp].positionSets[pp][npp].chromosomeID != -1)
1700 |                                                         {
1701 |                                                                 nextID = seqs[sp].positionSets[pp][npp].chromosomeID;
1702 |                                                                 nextOffset = seqs[sp].positionSets[pp][npp].chromosomeOffset;
1703 | 								nextID0 = seqs[sp0].positionSets[pp][npp].chromosomeID;
1704 | 								nextOffset0 = seqs[sp0].positionSets[pp][npp].chromosomeOffset;
1705 |                                                                 if(seqs[sp].positionSets[pp][npp].chromosomeID == seqs[sp].positionSets[pp][ppp].chromosomeID && npp - ppp < MAX)// small insertion
1706 |                                                                 {
1707 | 									if(nextOffset == chromosomeOffset + 1)
1708 | //                                                                      for(cpp = chromosomeOffset; cpp < nextOffset; cpp ++)
1709 |                                                                         {
1710 | //                                                                                p.nextID = nextID;
1711 | //                                                                                p.nextOffset = nextOffset;
1712 | //                                                                                p.nextItem = genome[p.nextID][p.nextOffset].nextPosition.size();
1713 | //                                                                                p.contigOffset = ppp;
1714 | //                                                                                genome[chromosomeID][chromosomeOffset].nextPosition.push_back(p);
1715 | //										cout << "SI" << endl;
1716 | 										if(s.size() > 0) s.clear();
1717 | 										for(np = ppp; np < ppp + length; np ++)
1718 | 											s.push_back(seqs[sp].nucleotides[np]);
1719 | 										if(nextS.size() > 0) nextS.clear();
1720 | 										//for(np = ppp + 1; np < ppp + length + 1; np ++)
1721 | 										size = npp + length < seqs[sp].nucleotides.size() ? npp + length : seqs[sp].nucleotides.size();
1722 | 										for(np = npp; np < size; np ++)
1723 | 											nextS.push_back(seqs[sp].nucleotides[np]);
1724 | 										updateKMer(genome, chromosomeID, chromosomeOffset, nextID, nextOffset, chromosomeID0, chromosomeOffset0, nextID0, nextOffset0, s, nextS, insertVariation, mrl);
1725 |                                                                                 //genome[chromosomeID][cpp].nextID = chromosomeID;
1726 |                                                                                 //genome[chromosomeID][cpp].nextOffset = cpp + 1;
1727 |                                                                         }
1728 | 
1729 | ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
1730 | 									else
1731 | 									{
1732 | 			                                                        if(s.size() > 0) s.clear();
1733 | 			                                                        for(np = ppp; np < ppp + length; np ++)
1734 | 			                                                                s.push_back(seqs[sp].nucleotides[np]);
1735 | 			                                                        if(nextS.size() > 0) nextS.clear();
1736 | 			                                                        updateKMer(genome, chromosomeID, chromosomeOffset, chromosomeID, chromosomeOffset + 1, chromosomeID0, chromosomeOffset0, -1, -1, s, nextS, insertVariation, mrl);
1737 | 			                                                        for(cpp = chromosomeOffset + 1; cpp < nextOffset - 1; cpp ++)
1738 | 			                                                        {
1739 | 			                                                                if(s.size() > 0) s.clear();
1740 | 			                                                                if(nextS.size() > 0) nextS.clear();
1741 | 			                                                                updateKMer(genome, chromosomeID, cpp, chromosomeID, cpp + 1, -1, -1, -1, -1, s, nextS, insertVariation, mrl);
1742 | 			                                                        }
1743 | 			                                                        if(s.size() > 0) s.clear();
1744 | 			                                                        if(nextS.size() > 0) nextS.clear();
1745 | 			                                                        //for(np = ppp + 1; np < ppp + length + 1; np ++)
1746 | 										size = npp + length < seqs[sp].nucleotides.size() ? npp + length : seqs[sp].nucleotides.size();
1747 | 										for(np = npp; np < size; np ++)
1748 | 			                                                                nextS.push_back(seqs[sp].nucleotides[np]);
1749 | 			                                                        updateKMer(genome, chromosomeID, cpp, chromosomeID, cpp + 1, -1, -1, nextID0, nextOffset0, s, nextS, insertVariation, mrl);
1750 | 
1751 | 									}
1752 | ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
1753 |                                                                 }
1754 |                                                                 else// large insertion
1755 | //large insertion is ignored for simplicity
1756 |                                                                 {
1757 | //                                                                      cout << "LI" << endl;
1758 |                                                                         p.nextID = chromosomeID;
1759 |                                                                         p.nextOffset = genome[chromosomeID].size();
1760 |                                                                         p.nextItem = 0;
1761 |                                                                         genome[chromosomeID][chromosomeOffset].contiMer.push_back(p);
1762 |                                                                         //genome[chromosomeID][chromosomeOffset].nextID = chromosomeID;
1763 |                                                                         //genome[chromosomeID][chromosomeOffset].nextOffset = genome[chromosomeID].size();
1764 |                                                                         for(i = 0; i < npp - ppp - 2; i ++)
1765 |                                                                         {
1766 |                                                                                 b.contiMer.clear();
1767 |                                                                                 b.nucleotide = seqs[sp].nucleotides[ppp + 1 + i];
1768 |                                                                                 p.nextID = chromosomeID;
1769 |                                                                                 p.nextOffset = genome[chromosomeID].size() + 1;
1770 |                                                                                 p.nextItem = 0;
1771 |                                                                                 b.contiMer.push_back(p);
1772 |                                                                                 //b.nextID = chromosomeID;
1773 |                                                                                 //b.nextOffset = genome[chromosomeID].size();
1774 |                                                                                 genome[chromosomeID].push_back(b);
1775 |                                                                         }
1776 |                                                                         b.contiMer.clear();
1777 |                                                                         b.nucleotide = seqs[sp].nucleotides[npp];
1778 |                                                                         p.nextID = nextID; //contigs[cp].positionSets[pp][npp].chromosomeID;
1779 |                                                                         p.nextOffset = nextOffset; //contigs[cp].positionSets[pp][npp].chromosomeOffset;
1780 |                                                                         p.nextItem = genome[p.nextID][p.nextOffset].contiMer.size();
1781 |                                                                         b.contiMer.push_back(p);
1782 |                                                                         genome[chromosomeID].push_back(b);
1783 |                                                                         //genome[chromosomeID][genome[chromosomeID].size() - 1].nextID = nextID;
1784 |                                                                         //genome[chromosomeID][genome[chromosomeID].size() - 1].nextOffset = nextOffset;
1785 |                                                                 }
1786 |                                                                 ppp = npp - 1;
1787 |                                                                 break;
1788 |                                                         }
1789 |                                                 }
1790 |                                         }
1791 |                                         else if(nextID == chromosomeID && nextOffset != chromosomeOffset + 1)// deletion from genome
1792 |                                         {
1793 |                                                 if(nextOffset - chromosomeOffset < SD)// small deletion
1794 |                                                 {
1795 | //                                                      cout << "SD" << endl;
1796 | 							if(s.size() > 0) s.clear();
1797 | 							for(np = ppp; np < ppp + length; np ++)
1798 | 								s.push_back(seqs[sp].nucleotides[np]);
1799 | 							if(nextS.size() > 0) nextS.clear();
1800 | //							for(np = ppp + 1; np < ppp + length + 1; np ++)
1801 | //								nextS.push_back(seqs[sp].nucleotides[np]);
1802 | 							updateKMer(genome, chromosomeID, chromosomeOffset, chromosomeID, chromosomeOffset + 1, chromosomeID0, chromosomeOffset0, -1, -1, s, nextS, insertVariation, mrl);
1803 |                                                         for(cpp = chromosomeOffset + 1; cpp < nextOffset - 1; cpp ++)
1804 |                                                         {
1805 | //                                                                p.nextID = chromosomeID;
1806 | //                                                                p.nextOffset = cpp + 1;
1807 | //                                                                p.nextItem = genome[p.nextID][p.nextOffset].nextPosition.size();
1808 | //                                                                genome[chromosomeID][cpp].nextPosition.push_back(p);
1809 | 								if(s.size() > 0) s.clear();
1810 | 								if(nextS.size() > 0) nextS.clear();
1811 | 								updateKMer(genome, chromosomeID, cpp, chromosomeID, cpp + 1, -1, -1, -1, -1, s, nextS, insertVariation, mrl);
1812 |                                                                 //genome[chromosomeID][cpp].nextID = chromosomeID;
1813 |                                                                 //genome[chromosomeID][cpp].nextOffset = cpp + 1;
1814 |                                                         }
1815 | 							if(s.size() > 0) s.clear();
1816 | 							if(nextS.size() > 0) nextS.clear();
1817 | 							for(np = ppp + 1; np < ppp + length + 1; np ++)
1818 | 								nextS.push_back(seqs[sp].nucleotides[np]);
1819 | 							updateKMer(genome, chromosomeID, cpp, chromosomeID, cpp + 1, -1, -1, nextID0, nextOffset0, s, nextS, insertVariation, mrl);
1820 | //The empty s and nextS make it possible to generate single nucleotide contig (the min contig size is not k-mer size)
1821 |                                                 }
1822 |                                                 else// large deletion
1823 |                                                 {
1824 | //                                                        cout << "LD" << endl;
1825 | //                                                        p.nextID = nextID;
1826 | //                                                        p.nextOffset = nextOffset;
1827 | //                                                        p.nextItem = genome[p.nextID][p.nextOffset].nextPosition.size();
1828 | //                                                        genome[chromosomeID][chromosomeOffset].nextPosition.push_back(p);
1829 | 							if(s.size() > 0) s.clear();
1830 | 							for(np = ppp; np < ppp + length; np ++)
1831 | 								s.push_back(seqs[sp].nucleotides[np]);
1832 | 							if(nextS.size() > 0) nextS.clear();
1833 | 							for(np = ppp + 1; np < ppp + length + 1; np ++)
1834 | 								nextS.push_back(seqs[sp].nucleotides[np]);
1835 | 							updateKMer(genome, chromosomeID, chromosomeOffset, nextID, nextOffset, chromosomeID0, chromosomeOffset0, nextID0, nextOffset0, s, nextS, insertVariation, mrl);
1836 |                                                         //genome[chromosomeID][chromosomeOffset].nextID = nextID;
1837 |                                                         //genome[chromosomeID][chromosomeOffset].nextOffset = nextOffset;
1838 | 
1839 |                                                 }
1840 |                                         }
1841 |                                         else// ordinary case
1842 |                                         {
1843 | //						  cout << "OD" << endl;
1844 | //                                                p.nextID = nextID;
1845 | //                                                p.nextOffset = nextOffset;
1846 | //                                                p.nextItem = genome[p.nextID][p.nextOffset].nextPosition.size();
1847 | //                                                genome[chromosomeID][chromosomeOffset].nextPosition.push_back(p);
1848 | 						if(s.size() > 0) s.clear();
1849 | 						for(np = ppp; np < ppp + length; np ++)
1850 | 							s.push_back(seqs[sp].nucleotides[np]);
1851 | 						if(nextS.size() > 0) nextS.clear();
1852 | 						for(np = ppp + 1; np < ppp + length + 1; np ++)
1853 | 							nextS.push_back(seqs[sp].nucleotides[np]);
1854 | 						updateKMer(genome, chromosomeID, chromosomeOffset, nextID, nextOffset, chromosomeID0, chromosomeOffset0, nextID0, nextOffset0, s, nextS, insertVariation, mrl);
1855 |                                                 //genome[chromosomeID][chromosomeOffset].nextID = nextID;
1856 |                                                 //genome[chromosomeID][chromosomeOffset].nextOffset = nextOffset;
1857 |                                         }
1858 |                                 }
1859 |                         }
1860 | 			if(tag2 == 1) switchSeqs(seqs[sp].positionSets[pp], seqs[sp0].positionSets[pp], seqs[sp].nucleotides, seqs[sp0].nucleotides);
1861 | 			if(tag0 == 1 && tag1 == 0) reverseComplement(seqs[sp].nucleotides);
1862 | 			else if(tag1 == 1 && tag0 == 0) reverseComplement(seqs[sp0].nucleotides);
1863 | 			else
1864 | 			{
1865 | 				cout << "UNKNOWN ERROR" << endl;
1866 | 				exit(-1);
1867 | 			}
1868 |                 }
1869 | 	}
1870 | }
1871 | 
1872 | void loadReadAlignment(vector<vector<Base> > & genome, int length, int insertVariation, int chromosomeID, int mrl)
1873 | { 
1874 | 	vector<Seq> reads;
1875 | 	ifstream r, ra;
1876 | 	int seqStartID, aliStartID, finish, rp;
1877 | 	string s;
1878 | //	int counter = 0;
1879 | 
1880 | 	r.open("tmp/_reads.fa");
1881 | 	s = "tmp/_reads_genome." + itoa(chromosomeID) + ".bowtie";
1882 | 	ra.open(s.c_str());
1883 | 	
1884 | 	seqStartID = aliStartID = -1;
1885 | cont:
1886 |         finish = loadSeq(r, reads, aliStartID, seqStartID);
1887 |         loadReadAli(ra, reads, aliStartID, seqStartID);
1888 |         updateGenomeWithRead(reads, genome, length, insertVariation, mrl);
1889 | 	if(finish == 0)// && counter < 2)
1890 | 	{
1891 | //		cout << counter ++ << endl;
1892 | 		reads.clear();
1893 | 		goto cont;
1894 | 	}
1895 | }
1896 | 
1897 | int contain(unsigned int startID1, unsigned int startOffset1, unsigned int endID1, unsigned int endOffset1, unsigned int startID2, unsigned int startOffset2, unsigned int endID2, unsigned int endOffset2)
1898 | {
1899 | 	if(startID1 == startID2 && endID1 == endID2 && startOffset1 <= startOffset2 && endOffset1 >= endOffset2)
1900 | 		return 1;
1901 | 	return 0;
1902 | }
1903 | 
1904 | void filterLowCoverage(vector<vector<Base> > & genome, int coverage)
1905 | {
1906 | 	unsigned int gp, cp, ip, cpp, pp;
1907 | 
1908 | 	for(gp = 0; gp < genome.size(); gp ++)
1909 | 		for(cp = 0; cp < genome[gp].size(); cp ++)
1910 | 			for(ip = 0; ip < genome[gp][cp].kMer.size(); ip ++)
1911 | 			{
1912 | 				if(genome[gp][cp].kMer[ip].contigID == -1)
1913 | 				{
1914 | 					if(genome[gp][cp].kMer[ip].coverage < coverage)
1915 | 						genome[gp][cp].kMer[ip].traversed = 1;
1916 | 				}
1917 | 			}
1918 | }
1919 | 
1920 | void countBranches(vector<vector<Base> > & genome)
1921 | {
1922 | 	unsigned int gp, cp, ip, nBranches = 0;
1923 | 
1924 | 	for(gp = 0; gp < genome.size(); gp ++)
1925 | 		for(cp = 0; cp < genome[gp].size(); cp ++)
1926 | 			for(ip = 0; ip < genome[gp][cp].kMer.size(); ip ++)
1927 | 				if(genome[gp][cp].kMer[ip].traversed == 0 && genome[gp][cp].kMer[ip].next.size() > 1)
1928 | 					nBranches ++;
1929 | 
1930 | 	cout << "ALERT: " << nBranches << " branches in de Bruijn graph" << endl;
1931 | }
1932 | 
1933 | void checkContiMers(vector<vector<Base> > & genome)
1934 | {
1935 |         unsigned int gp, cp, ip, nBranches = 0;
1936 | 
1937 |         for(gp = 0; gp < genome.size(); gp ++)
1938 |                 for(cp = 0; cp < genome[gp].size(); cp ++)
1939 |                         for(ip = 0; ip < genome[gp][cp].contiMer.size(); ip ++)
1940 |                                 if(genome[gp][cp].contiMer[ip].contigID != -1 && genome[gp][cp].contiMer[ip].contigOffset == -1)
1941 |                                         cout << "ERROR@ " << gp << ", " << cp << ", " << ip << endl;
1942 | }
1943 | 
1944 | char max(int a, int c, int g, int t, int n)
1945 | {
1946 | 	if(a == 0 && c == 0 && g == 0 && t == 0 && n == 0) return 'X';
1947 | 	if(a >= c && a >= g && a >= t && a >= n) return 'A';
1948 | 	if(c >= a && c >= g && c >= t && c >= n) return 'C';
1949 | 	if(g >= a && g >= c && g >= t && g >= n) return 'G';
1950 | 	if(t >= a && t >= c && t >= g && t >= n) return 'T';
1951 | 	if(n >= a && n >= c && n >= g && n >= t) return 'N';
1952 | }
1953 | 
1954 | void extdContigs1(vector<vector<Base> > & genome, int coverage, int k, int chromosomeID)
1955 | {
1956 | 	unsigned int gp, cp, ip, gpp, cpp, ipp, gppBak, cppBak, ippBak, seqID, i, kMerTag, numItem, nextItem, sp, gpBak, cpBak, ipBak, np, gpp0, cpp0, ngp, ncp, nip, ippp, item, next, count, nCount, startIDBak = -1, startOffsetBak = -1, endIDBak = -1, endOffsetBak = -1;
1957 | 	Contig contig;
1958 | 	vector<char> sBak;
1959 | 	vector<Contig> contigsBuf;
1960 | 	char nucleotide;
1961 | 
1962 | //******************************************************************************************************************************************************************************
1963 | 	ofstream out;
1964 |         string s = "tmp/_pre_extended_contigs." + itoa(chromosomeID) + ".fa";
1965 |         out.open(s.c_str());
1966 | //******************************************************************************************************************************************************************************
1967 | 
1968 | 	filterLowCoverage(genome, coverage);
1969 | //	countBranches(genome);
1970 | 
1971 | 	seqID = 0;
1972 | 	for(gp = 0; gp < genome.size(); gp ++)
1973 | 		for(cp = 0; cp < genome[gp].size();)
1974 | 		{
1975 | //			if(genome[gp][cp].kMer.size() != 0)
1976 | 				for(ip = 0; ip < genome[gp][cp].kMer.size(); ip ++)
1977 | 				{
1978 | 					if(genome[gp][cp].kMer[ip].traversed == 0)
1979 | 					{
1980 | 						gpp = gp;
1981 | 						cpp = cp;
1982 | 						ipp = ip;
1983 | 						kMerTag = 1;
1984 | 						contigs.push_back(contig);
1985 | 						contigs[contigs.size() - 1].startID = gp;
1986 | 						contigs[contigs.size() - 1].startOffset = cp;
1987 | 						contigs[contigs.size() - 1].startID0 = genome[gp][cp].kMer[ip].chromosomeID0;
1988 | 						contigs[contigs.size() - 1].startOffset0 = genome[gp][cp].kMer[ip].chromosomeOffset0;
1989 | 						contigs[contigs.size() - 1].extended = 0;
1990 | 
1991 | 						while(kMerTag == 1 && genome[gpp][cpp].kMer[ipp].traversed == 0 || kMerTag == 0)
1992 | 						{
1993 | 							if(kMerTag == 0)
1994 | 								contigs[contigs.size() - 1].nucleotides.push_back(genome[gpp][cpp].contiMer[ipp].nucleotide);
1995 | 							else
1996 | 							{
1997 | 								nucleotide = max(genome[gpp][cpp].kMer[ipp].A, genome[gpp][cpp].kMer[ipp].C, genome[gpp][cpp].kMer[ipp].G, genome[gpp][cpp].kMer[ipp].T, genome[gpp][cpp].kMer[ipp].N);
1998 | 								if(nucleotide != 'X')
1999 | 									contigs[contigs.size() - 1].nucleotides.push_back(nucleotide);
2000 | 								else
2001 | 									contigs[contigs.size() - 1].nucleotides.push_back(genome[gpp][cpp].nucleotide);
2002 | 							}
2003 | 
2004 | 							if(kMerTag == 1 && genome[gpp][cpp].kMer[ipp].contigOffset != -1 || kMerTag == 0) 
2005 | 							//incorporate novel ones
2006 | 								contigs[contigs.size() - 1].extended = 1;
2007 | 
2008 | 							if(kMerTag == 1)
2009 | 							{
2010 | 								gpp0 = genome[gpp][cpp].kMer[ipp].chromosomeID0;
2011 | 								cpp0 = genome[gpp][cpp].kMer[ipp].chromosomeOffset0;
2012 | //								cout << "k-mer " << cpp << ": "  << genome[gpp][cpp].nucleotide << " | ";
2013 | 								genome[gpp][cpp].kMer[ipp].traversed = 1;
2014 | 								gpBak = gpp;
2015 | 								cpBak = cpp;
2016 | 								ipBak = ipp;
2017 | 								sBak = genome[gpp][cpp].kMer[ipp].s;
2018 | ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
2019 | 
2020 | 								nCount = 0;
2021 | 								next = -1;
2022 | 								for(np = 0; np < genome[gpp][cpp].kMer[ipp].next.size(); np ++)
2023 | 								{
2024 | 									ngp = genome[gpp][cpp].kMer[ipp].next[np].nextID;
2025 | 									ncp = genome[gpp][cpp].kMer[ipp].next[np].nextOffset;
2026 | 									nip = genome[gpp][cpp].kMer[ipp].next[np].nextItem;
2027 | 									if(ngp != -1 && genome[ngp][ncp].kMer[nip].traversed == 0)
2028 | 									{
2029 | 										next = np;
2030 | 										nCount ++;
2031 | 									}
2032 | 								}
2033 | 								if(nCount == 1)
2034 | 
2035 | ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
2036 | //								if(genome[gpp][cpp].kMer[ipp].next.size() == 1 && genome[gpp][cpp].kMer[ipp].next[0].nextID != -1)// need improvement considering filteration
2037 | 								{
2038 | 									gppBak = genome[gpp][cpp].kMer[ipp].next[next].nextID;
2039 | 									cppBak = genome[gpp][cpp].kMer[ipp].next[next].nextOffset;
2040 | 									ippBak = genome[gpp][cpp].kMer[ipp].next[next].nextItem;
2041 | 									gpp = gppBak;
2042 | 									cpp = cppBak;
2043 | 									ipp = ippBak;
2044 | 									kMerTag = 1;
2045 | //									cout << "keep k-mer" << " | ";
2046 | 								}
2047 | 								else if(genome[gpp][cpp].contiMer.size() == 1 && genome[gpp][cpp].contiMer[0].nextID != -1)
2048 | 								{
2049 | 									gppBak = genome[gpp][cpp].contiMer[0].nextID;
2050 | 									cppBak = genome[gpp][cpp].contiMer[0].nextOffset;
2051 | 									ippBak = genome[gpp][cpp].contiMer[0].nextItem;
2052 | 									gpp = gppBak;
2053 | 									cpp = cppBak;
2054 | 									ipp = ippBak;
2055 | 									kMerTag = 0;
2056 | //									cout << "switch from k-mer to conti-mer" << " | ";
2057 | 								}
2058 | 								else
2059 | 									kMerTag = -1;
2060 | 							}
2061 | 							else// if(kMerTag == 0)
2062 | 							{
2063 | //								cout << "conti-mer " << cpp << ": " << genome[gpp][cpp].nucleotide << " | ";
2064 | 								if(genome[gpp][cpp].contiMer[ipp].nextID != -1)
2065 | 								{
2066 | 									gppBak = genome[gpp][cpp].contiMer[ipp].nextID;
2067 | 									cppBak = genome[gpp][cpp].contiMer[ipp].nextOffset;
2068 | 									ippBak = genome[gpp][cpp].contiMer[ipp].nextItem;
2069 | 									gpp = gppBak;
2070 | 									cpp = cppBak;
2071 | 									ipp = ippBak;
2072 | 									kMerTag = 0;
2073 | //maybe not necessary to get back to k-mer?
2074 | //									for(ippBak = 0, numItem = 0; ippBak < genome[gpp][cpp].kMer.size(); ippBak ++)
2075 | //										if(genome[gpp][cpp].kMer[ippBak].traversed == 0)
2076 | //										{
2077 | //											numItem ++;
2078 | //											nextItem = ippBak;
2079 | //										}
2080 | //									if(numItem == 1)
2081 | //									{
2082 | //										cout << "switch from conti-mer to k-mer (1)" << " | ";
2083 | //										ipp = nextItem;
2084 | //										kMerTag = 1;
2085 | //									}
2086 | //									else
2087 | //										cout << "keep conti-mer" << " | ";
2088 | 								}
2089 | 								else
2090 | 								{
2091 | ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
2092 | 
2093 | 									count = nCount = 0;
2094 | 									item = next = -1;
2095 | 									for(ippp = 0; ippp < genome[gpp][cpp].kMer.size(); ippp ++)
2096 | 										if(genome[gpp][cpp].kMer[ippp].traversed == 0)
2097 | 										{
2098 | 											count ++;
2099 | 											item = ippp;
2100 | 										}
2101 | 									if(count == 1)
2102 | 									{
2103 | 										for(np = 0; np < genome[gpp][cpp].kMer[item].next.size(); np ++)
2104 | 										{
2105 | 											ngp = genome[gpp][cpp].kMer[item].next[np].nextID;
2106 | 											ncp = genome[gpp][cpp].kMer[item].next[np].nextOffset;
2107 | 											nip = genome[gpp][cpp].kMer[item].next[np].nextItem;
2108 | 											if(ngp != -1 && genome[ngp][ncp].kMer[nip].traversed == 0)
2109 | 											{
2110 | 												nCount ++;
2111 | 												next = np;
2112 | 											}
2113 | 										}
2114 | 									}
2115 |                                                                 	if(nCount == 1)
2116 | 
2117 | ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
2118 | //									if(genome[gpp][cpp].kMer.size() == 1 && genome[gpp][cpp].kMer[0].next.size() == 1 && genome[gpp][cpp].kMer[0].next[0].nextID != -1)
2119 | 									{
2120 | 										gppBak = genome[gpp][cpp].kMer[item].next[next].nextID;
2121 | 										cppBak = genome[gpp][cpp].kMer[item].next[next].nextOffset;
2122 | 										ippBak = genome[gpp][cpp].kMer[item].next[next].nextItem;
2123 | 										gpp = gppBak;
2124 | 										cpp = cppBak;
2125 | 										ipp = ippBak;
2126 | ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
2127 | 										if(genome[gpp][cpp].kMer[ipp].traversed == 0)
2128 | 											kMerTag = 1;
2129 | 										else
2130 | 											kMerTag = -2;
2131 | ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
2132 | 										//kMerTag = 1;
2133 | //										cout << "switch from conti-mer to k-mer (2)" << " | ";
2134 | 									}
2135 | 									else
2136 | 										kMerTag = -2;
2137 | 								}
2138 | 							}
2139 | 						}
2140 | 
2141 | //xpp and xppBak are always the same
2142 | 						if(kMerTag == 1)// && genome[gpp][cpp].kMer[ipp].traversed == 1
2143 | 						{
2144 | 							contigs[contigs.size() - 1].endID = gppBak;
2145 | 							contigs[contigs.size() - 1].endOffset = cppBak;
2146 | 						}
2147 | 						else//kMerTag == -1 || kMerTag == -2
2148 | 						{
2149 |                                                 	contigs[contigs.size() - 1].endID = gpp;
2150 |                                                 	contigs[contigs.size() - 1].endOffset = cpp;
2151 | 						}
2152 | 
2153 | 						if(kMerTag == 1 || kMerTag == -1)
2154 | 						{
2155 |                                                 	contigs[contigs.size() - 1].endID0 = genome[gpp][cpp].kMer[ipp].chromosomeID0;
2156 |                                                 	contigs[contigs.size() - 1].endOffset0 = genome[gpp][cpp].kMer[ipp].chromosomeOffset0;
2157 | 						}
2158 | 						else// kMerTag == -2
2159 | 						{
2160 | 							contigs[contigs.size() - 1].endID0 = -1;
2161 | 							contigs[contigs.size() - 1].endOffset0 = -1;
2162 | 						}
2163 | 
2164 | 						if(kMerTag == -1 || kMerTag == 1)
2165 | 						{
2166 | //append the last |s| - 1 bases only if getting out of while from k-mer rather than conti-mer
2167 | 							for(sp = 1; sp < sBak.size(); sp ++)
2168 | 								contigs[contigs.size() - 1].nucleotides.push_back(sBak[sp]);
2169 | ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
2170 | 							contigs[contigs.size() - 1].endOffset = contigs[contigs.size() - 1].endOffset + sBak.size() - 1;
2171 | 							contigs[contigs.size() - 1].endOffset0 = contigs[contigs.size() - 1].endOffset0 + sBak.size() - 1;
2172 | ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
2173 | 						}
2174 | 
2175 | //******************************************************************************************************************************************************************************
2176 | 						if(/*contigs[contigs.size() - 1].extended == 1 && */ contain(startIDBak, startOffsetBak, endIDBak, endOffsetBak, contigs[contigs.size() - 1].startID, contigs[contigs.size() - 1].startOffset, contigs[contigs.size() - 1].endID, contigs[contigs.size() - 1].endOffset) == 0)
2177 | 						{
2178 | 							out << ">" << seqID ++ << ", " /*<< contigs[contigs.size() - 1].contigID << ", " << contigs[contigs.size() - 1].contigOffset << ", " << contigs[contigs.size() - 1].contigID0 << ", " << contigs[contigs.size() - 1].contigOffset0 << ", "*/ << contigs[contigs.size() - 1].extended << ", " << contigs[contigs.size() - 1].startID << ", " << contigs[contigs.size() - 1].startOffset << ", " << contigs[contigs.size() - 1].endID << ", " << contigs[contigs.size() - 1].endOffset << ", " << contigs[contigs.size() - 1].startID0 << ", " << contigs[contigs.size() - 1].startOffset0 << ", " << contigs[contigs.size() - 1].endID0 << ", " << contigs[contigs.size() - 1].endOffset0 << " " << endl;
2179 | 							for(i = 0; i < contigs[contigs.size() - 1].nucleotides.size(); i ++)
2180 | 							{
2181 | 								out << contigs[contigs.size() - 1].nucleotides[i];
2182 | 								if((i + 1) % 60 == 0 || i == contigs[contigs.size() - 1].nucleotides.size() - 1)
2183 | 									out << endl;
2184 | 							}
2185 | 							startIDBak = contigs[contigs.size() - 1].startID;
2186 | 							startOffsetBak = contigs[contigs.size() - 1].startOffset;
2187 | 							endIDBak = contigs[contigs.size() - 1].endID;
2188 | 							endOffsetBak = contigs[contigs.size() - 1].endOffset;
2189 | 						}
2190 | 						contigs.clear();
2191 | //******************************************************************************************************************************************************************************
2192 | 					}
2193 | 				}
2194 | 			if(endOffsetBak - startOffsetBak > 100000)
2195 | 			{
2196 | 				if(gp == endIDBak && cp + 1000 < endOffsetBak)
2197 | 					cp = cp + 1000;
2198 | 				else
2199 | 					cp ++;
2200 | 			}
2201 | 			else
2202 | 				cp ++;
2203 | 		}
2204 | }
2205 | 
2206 | void parse(string buf, int & extended, unsigned int & startID, unsigned int & startOffset, unsigned int & endID, unsigned int & endOffset, unsigned int & startID0, unsigned int & startOffset0, unsigned int & endID0, unsigned int & endOffset0)
2207 | {
2208 | 	int i, item = 0, j1 = 0, j2 = 0, j3 = 0, j4 = 0, j5 = 0, j6 = 0, j7 = 0, j8 = 0, j9 = 0, j10 = 0, j11 = 0, j12 = 0, j13 = 0;
2209 | 	char contigIDBuf[20] = {'\0'}, contigOffsetBuf[20] = {'\0'}, contigID0Buf[20] = {'\0'}, contigOffset0Buf[20] = {'\0'}, extendedBuf[20] = {'\0'}, startIDBuf[20] = {'\0'}, startOffsetBuf[20] = {'\0'}, endIDBuf[20] = {'\0'}, endOffsetBuf[20] = {'\0'}, startID0Buf[20] = {'\0'}, startOffset0Buf[20] = {'\0'}, endID0Buf[20] = {'\0'}, endOffset0Buf[20] = {'\0'};
2210 | 
2211 | 	for(i = 0; i < buf.size(); i ++)
2212 | 	{
2213 | 		if(buf[i] == ' ')
2214 | 		{
2215 | 			item ++;
2216 | 			continue;
2217 | 		}
2218 | 
2219 | 		if(item == 0)
2220 | 			continue;
2221 | 		else if(item == 1)
2222 | 			extendedBuf[j1 ++] = buf[i];
2223 | 		else if(item == 2)
2224 | 			startIDBuf[j2 ++] = buf[i];
2225 | 		else if(item == 3)
2226 | 			startOffsetBuf[j3 ++] = buf[i];
2227 | 		else if(item == 4)
2228 | 			endIDBuf[j4 ++] = buf[i];
2229 | 		else if(item == 5)
2230 | 			endOffsetBuf[j5 ++] = buf[i];
2231 |                 else if(item == 6)
2232 |                         startID0Buf[j6 ++] = buf[i];
2233 |                 else if(item == 7)
2234 |                         startOffset0Buf[j7 ++] = buf[i];
2235 |                 else if(item == 8)
2236 |                         endID0Buf[j8 ++] = buf[i];
2237 |                 else if(item == 9)
2238 |                         endOffset0Buf[j9 ++] = buf[i];
2239 | 		else if(item == 10)
2240 | 			break;
2241 | 		else
2242 | 		{
2243 | 			cout << "UNKNOWN ERROR" << endl;
2244 | 			exit(-1);
2245 | 		}
2246 | 	}
2247 | 	extended = atoi(extendedBuf);
2248 | 	startID = atoi(startIDBuf);
2249 | 	startOffset = atoi(startOffsetBuf);
2250 | 	endID = atoi(endIDBuf);
2251 | 	endOffset = atoi(endOffsetBuf);
2252 |         startID0 = atoi(startID0Buf);
2253 |         startOffset0 = atoi(startOffset0Buf);
2254 |         endID0 = atoi(endID0Buf);
2255 |         endOffset0 = atoi(endOffset0Buf);
2256 | }
2257 | 
2258 | void loadContigs(int chromosomeID)
2259 | {
2260 | 	int cp, cpp, i;
2261 | 	ifstream in;
2262 | 	string buf;
2263 | 	Contig contig;
2264 | 
2265 | 	string s = "tmp/_pre_extended_contigs." + itoa(chromosomeID) + ".fa";
2266 |         in.open(s.c_str());
2267 | 	contig.extended = /*contig.contigID = contig.contigOffset = contig.contigID0 = contig.contigOffset0 =*/ contig.startID = contig.startOffset = contig.endID = contig.endOffset = contig.startID0 = contig.startOffset0 = contig.endID0 = contig.endOffset0 = -1;
2268 | 
2269 | 	if(in.is_open())
2270 | 	{
2271 | 		while(in.good())
2272 | 		{
2273 | 			getline(in, buf);
2274 | 			if(buf[0] == 0)
2275 | 				break;
2276 | 
2277 | 			if(buf[0] == '>')
2278 | 			{
2279 | 				contigs.push_back(contig);
2280 | 				parse(buf, contigs[contigs.size() - 1].extended, /*contigs[contigs.size() - 1].contigID, contigs[contigs.size() - 1].contigOffset, contigs[contigs.size() - 1].contigID0, contigs[contigs.size() - 1].contigOffset0,*/ contigs[contigs.size() - 1].startID, contigs[contigs.size() - 1].startOffset, contigs[contigs.size() - 1].endID, contigs[contigs.size() - 1].endOffset, contigs[contigs.size() - 1].startID0, contigs[contigs.size() - 1].startOffset0, contigs[contigs.size() - 1].endID0, contigs[contigs.size() - 1].endOffset0);
2281 | 			}
2282 | 			else
2283 | 			{
2284 | 				for(i = 0; i < buf.size(); i ++)
2285 | 					contigs[contigs.size() - 1].nucleotides.push_back(buf[i]);
2286 | 			}
2287 | 		}
2288 | 	}
2289 | 	else
2290 | 	{
2291 | 		cout << "CANNOT OPEN FILE!" << endl;
2292 | 		exit(-1);
2293 | 	}
2294 | }
2295 | 
2296 | void extdContigs2(int chromosomeID)
2297 | {
2298 | 	int cp, cpp, cppBak, np;
2299 | 	vector<Contig> contigsBuf;
2300 | 
2301 | 	loadContigs(chromosomeID);
2302 | 
2303 | 	for(cp = 0; cp < contigs.size(); cp ++)
2304 | 	{
2305 | 		if(contigs[cp].extended == 1)
2306 | 		{
2307 | 			for(cpp = cp + 1; cpp < contigs.size(); cpp ++)
2308 | 			//if(contigs[cpp].extended == 1)
2309 | 			{
2310 | 				if(contain(contigs[cp].startID, contigs[cp].startOffset, contigs[cp].endID, contigs[cp].endOffset,
2311 | 				contigs[cpp].startID, contigs[cpp].startOffset, contigs[cpp].endID, contigs[cpp].endOffset))
2312 | 				{
2313 | //					cout << "duplication removed" << endl;
2314 | 					contigs[cpp].extended = 2;
2315 | 				}
2316 | 				else if(contigs[cp].endID != contigs[cpp].startID || contigs[cp].endOffset < contigs[cpp].startOffset)
2317 | 					break;
2318 | 			}
2319 | 		}
2320 | 	}
2321 | 
2322 | 	for(cp = contigs.size() - 1; cp != -1; cp --)
2323 | 	{
2324 | 		if(contigs[cp].extended == 1)
2325 | 		{
2326 | 			for(cpp = cp - 1; cpp != -1; cpp --)
2327 | 			//if(contigs[cpp].extended == 1)
2328 | 			{
2329 | 				if(contain(contigs[cp].startID, contigs[cp].startOffset, contigs[cp].endID, contigs[cp].endOffset,
2330 |                                 contigs[cpp].startID, contigs[cpp].startOffset, contigs[cpp].endID, contigs[cpp].endOffset))
2331 | 				{
2332 | //					cout << "duplication removed" << endl;
2333 | 					contigs[cpp].extended = 2;
2334 | 				}
2335 | 				else if(contigs[cpp].endID != contigs[cp].startID || contigs[cpp].endOffset < contigs[cp].startOffset)
2336 | 					break;
2337 | 			}
2338 | 		}
2339 | 	}
2340 | //remove duplication (0: no contig; 1: extended with contig; 2: filtered)
2341 | 
2342 | 	for(cp = 0; cp < contigs.size(); cp ++)
2343 | 	{
2344 | cont:
2345 | 		if(contigs[cp].extended == 1)
2346 | 		{
2347 | 			cppBak = -1;
2348 | 			contigsBuf.clear();
2349 | 			for(cpp = cp + 1; cpp < contigs.size(); cpp ++)
2350 | 				if(contigs[cpp].extended != 2)
2351 | 				{
2352 | 					if(//contigs[cp].endID == contigs[cpp].startID && 
2353 | 					//contigs[cp].endOffset < contigs[cpp].startOffset && 
2354 | 					contigs[cp].endOffset >= contigs[cpp].startOffset)
2355 | 					{
2356 | //						cout << "potential extension: " << contigs[cp].startOffset << " to " << contigs[cpp].startOffset << endl;
2357 | 						contigsBuf.push_back(contigs[cpp]);
2358 | 						cppBak = cpp;
2359 | 					}
2360 | 					else if(//contigs[cp].endID != contigs[cpp].startID || 
2361 | 					contigs[cp].endOffset < contigs[cpp].startOffset)
2362 | 						break;
2363 | 				}
2364 | 			if(contigsBuf.size() == 1)
2365 | 			{
2366 | //				cout << "real extension: " << contigs[cp].startOffset << " to " <<  contigs[cppBak].startOffset << endl;
2367 | 				contigs[cppBak].extended = 2; 
2368 | 				for(np = //contigs[cp].endOffset < contigs[cpp].startOffset ? 0 : 
2369 | 					contigs[cp].endOffset - contigsBuf[0].startOffset + 1; np < contigsBuf[0].nucleotides.size(); np ++)
2370 | 					contigs[cp].nucleotides.push_back(contigsBuf[0].nucleotides[np]);
2371 | 				contigs[cp].endID = contigsBuf[0].endID;
2372 |                                 contigs[cp].endOffset = contigsBuf[0].endOffset;
2373 |                                 contigs[cp].endID0 = contigsBuf[0].endID0;
2374 |                                 contigs[cp].endOffset0 = contigsBuf[0].endOffset0;
2375 | 				goto cont;
2376 | 			}
2377 | 		}
2378 | 	}
2379 | //join contigs
2380 | }
2381 | 
2382 | void extendContigs(vector<vector<Base> > & genome, int coverage, int k, int chromosomeID)
2383 | {
2384 | 	extdContigs1(genome, coverage, k, chromosomeID);
2385 |         extdContigs2(chromosomeID);
2386 | }
2387 | 
2388 | int overlap(unsigned int x1, unsigned int y1, unsigned int x2, unsigned int y2) 
2389 | {
2390 | 	if(x1 <= x2 && x2 <= y1 && y1 <= y2 && (int)y1 - (int)x2 > 0 || x2 <= x1 && x1 <= y2 && y2 <= y1 && (int)y2 - (int)x1 > 0 || x1 <= x2 && x2 <= y2 && y2 <= y1 && (int)y2 - (int)x2 > 0 || x2 <= x1 && x1 <= y1 && y1 <= y2 && (int)y1 - (int)x1 > 0)
2391 | 		return 1;
2392 | 	else
2393 | 		return 0;
2394 | }
2395 | 
2396 | void scaffoldContigs(vector<vector<Base> > & genome, int chromosomeID)
2397 | {
2398 | 	unsigned int seqID, cp, cp0, cpp, i, sp, spp;
2399 | 	vector<vector<char> > scaffolds;
2400 | 	vector<char> scaffold;
2401 | 	int cont, covered;
2402 | 
2403 | 	seqID = 0;
2404 | 	for(cp = 0; cp < contigs.size(); cp ++)
2405 | 	{
2406 | 		if(contigs[cp].startID != -1 && contigs[cp].extended == 1)
2407 | 		{
2408 | 			scaffolds.push_back(scaffold);
2409 | 			for(cpp = 0; cpp < contigs[cp].nucleotides.size(); cpp ++)
2410 | 				scaffolds[scaffolds.size() - 1].push_back(contigs[cp].nucleotides[cpp]);
2411 | 			contigs[cp].startID = -1;
2412 | 		}
2413 | 		else
2414 | 			continue;
2415 | 
2416 | 		cont = 1;
2417 | 		while(contigs[cp].startID0 == contigs[cp].endID0 && cont)
2418 | 		{
2419 | 			cont = 0;
2420 | 			for(cp0 = cp + 1; cp0 < contigs.size(); cp0 ++)
2421 | 			{
2422 | 				if(cp0 != cp && contigs[cp].endID0 == contigs[cp0].startID && contigs[cp0].startID == contigs[cp0].endID && overlap(contigs[cp].startOffset0, contigs[cp].endOffset0, contigs[cp0].startOffset, contigs[cp0].endOffset) && contigs[cp0].extended == 1)
2423 | 				{
2424 | 					if(contigs[cp0].startOffset > contigs[cp].endOffset)
2425 | 					{
2426 | 						covered = 0;
2427 | 						for(i = 0; i < contigs[cp0].startOffset - contigs[cp].endOffset - 1; i ++)
2428 | 							if(genome[0][contigs[cp].endOffset + i + 1].kMer.size() > 0 || genome[0][contigs[cp].endOffset + i + 1].contiMer.size() > 0)
2429 | 								covered ++;
2430 | 						if(contigs[cp0].startOffset - contigs[cp].endOffset - 1 != 0 && (double) covered / (contigs[cp0].startOffset - contigs[cp].endOffset - 1) >= 0.5 || contigs[cp0].startOffset - contigs[cp].endOffset - 1 == 0)
2431 | 							for(i = 0; i < contigs[cp0].startOffset - contigs[cp].endOffset - 1; i ++)
2432 | 								scaffolds[scaffolds.size() - 1].push_back(genome[0][contigs[cp].endOffset + i + 1].nucleotide);
2433 | //								scaffolds[scaffolds.size() - 1].push_back('N');
2434 | 						else
2435 | 							goto conti;
2436 | 					}
2437 | 					for(cpp = 0; cpp < contigs[cp0].nucleotides.size(); cpp ++)
2438 | 						scaffolds[scaffolds.size() - 1].push_back(contigs[cp0].nucleotides[cpp]);
2439 | 					contigs[cp0].startID = -1;
2440 | 					cp = cp0;
2441 | 					cont = 1;
2442 | 					break;
2443 | 				}
2444 | conti:;
2445 | 			}
2446 | 		}	
2447 | 
2448 | 	}
2449 | 
2450 | //output extended contigs
2451 | 	ofstream out;
2452 | 	string s = "tmp/_extended_contigs." + itoa(chromosomeID) + ".fa";
2453 | 	out.open(s.c_str());
2454 | 	for(sp = 0; sp < scaffolds.size(); sp ++)
2455 | 	{
2456 | 		out << ">" << seqID ++ << endl;
2457 | 		for(spp = 0; spp < scaffolds[sp].size(); spp ++)
2458 | 		{
2459 | 			out << scaffolds[sp][spp];
2460 | 			if((spp + 1) % 60 == 0 || spp == scaffolds[sp].size() - 1)
2461 | 				out << endl;
2462 | 		}
2463 | 	}
2464 | }
2465 | 
2466 | int readLog()
2467 | {
2468 |         ifstream in;
2469 |         string buf;
2470 | 
2471 |         in.open("tmp/log.txt");
2472 |         if(in.is_open())
2473 |         {
2474 |                 getline(in, buf);
2475 |                 return(atoi(buf.c_str()));
2476 |         }
2477 |         else
2478 |         {
2479 |                 cout << "CANNOT OPEN FILE!" << endl;
2480 |                 exit(-1);
2481 |         }
2482 | 
2483 | }
2484 | 
2485 | void ref1()
2486 | {
2487 | 	int numChromosomes, i, j, k, seqID;
2488 | 	string s;
2489 | 	vector<vector<char> > initContigs, extdContigs;
2490 | 	vector<char> contig;
2491 | 	string buf;
2492 | 	unsigned int sourceID, targetID, targetStart, targetEnd, targetGap, sourceStart, sourceEnd, sourceGap, sourceSize, targetSize, fr;
2493 | 	vector<int> initTags, extdTags;
2494 | 	ifstream in, ex, ps;
2495 | 	vector<Segment> seg;
2496 | 	ofstream e, ini;
2497 | 	vector<int> initNums;
2498 | 
2499 | 	numChromosomes = readLog();
2500 | 
2501 | ///////////////////////////////////////////////////////////////////for easy alignment/////////////////////////////////////////////////////////////////////////////////////
2502 | 
2503 | 	ifstream tmpIn;
2504 | 	ofstream tmpOut;
2505 | 	vector<vector<char> > tmpContigs;
2506 | 	for(i = 0; i < numChromosomes; i ++)
2507 | 	{
2508 | 		tmpIn.clear();
2509 | 		tmpOut.clear();
2510 | 		tmpContigs.clear();
2511 | 		seqID = 0;
2512 | 
2513 | 		s = "tmp/_initial_contigs." + itoa(i) + ".fa";
2514 | 		tmpIn.open(s.c_str());
2515 | 		s = "tmp/_short_initial_contigs." + itoa(i) + ".fa";
2516 | 		tmpOut.open(s.c_str());
2517 | 		if(tmpIn.is_open())
2518 | 		{
2519 | 			while(tmpIn.good())
2520 | 			{
2521 | 				getline(tmpIn, buf);
2522 | 				if(buf[0] == 0) break;
2523 | 
2524 | 				if(buf[0] == '>')
2525 | 					tmpContigs.push_back(contig);
2526 | 				else
2527 | 					for(j = 0; j < buf.size(); j ++)
2528 | 						tmpContigs[tmpContigs.size() - 1].push_back(buf[j]);
2529 | 			}
2530 | 		}
2531 | 		else
2532 | 		{
2533 | 			cout << "CANNOT OPEN FILE!" << endl;
2534 | 			exit(-1);
2535 | 		}
2536 | 
2537 | 		for(i = 0; i < tmpContigs.size(); i ++)
2538 | 		{
2539 | 			tmpOut << ">" << i << endl;
2540 | 			if(tmpContigs[i].size() > SMALL_CHUNK)
2541 | 			{
2542 | 				for(j = 0; j < SMALL_CHUNK; j ++)
2543 | 				{
2544 | 					tmpOut << tmpContigs[i][j];
2545 | 					if((j + 1) % 60 == 0 || j == SMALL_CHUNK - 1)
2546 | 						tmpOut << endl;
2547 | 				}
2548 | 			}
2549 | 			else
2550 | 			{
2551 | 				for(j = 0; j < tmpContigs[i].size(); j ++)
2552 | 				{
2553 | 					tmpOut << tmpContigs[i][j];
2554 | 					if((j + 1) % 60 == 0 || j == tmpContigs[i].size() - 1)
2555 | 						tmpOut << endl;
2556 | 				}
2557 | 			}
2558 | 		}
2559 | 
2560 | 		tmpIn.close();
2561 | 		tmpOut.close();
2562 | 	}
2563 | 
2564 | //////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
2565 |         for(i = 0; i < numChromosomes; i ++)
2566 | 	{
2567 | 		s = "pblat tmp/_extended_contigs." + itoa(i) + ".fa tmp/_short_initial_contigs." + itoa(i) + ".fa -noHead tmp/_short_initial_contigs_extended_contigs." + itoa(i) + ".psl -threads=8 > blat_doc.txt 2> blat_doc.txt";
2568 | 		if(system(s.c_str()) != 0)
2569 | 		{
2570 | 			s = "blat tmp/_extended_contigs." + itoa(i) + ".fa tmp/_short_initial_contigs." + itoa(i) + ".fa -noHead tmp/_short_initial_contigs_extended_contigs." + itoa(i) + ".psl > blat_doc.txt 2> blat_doc.txt";
2571 | 			if(system(s.c_str()) != 0) {cout << "BLAT CALL FAILED!" << endl; exit(-1);}
2572 | 		}
2573 | 	}
2574 | 
2575 | 	seqID = 0;
2576 | 	for(i = 0; i < numChromosomes; i ++)
2577 | 	{
2578 | 		s = "tmp/_extended_contigs." + itoa(i) + ".fa";
2579 | 		ex.clear();
2580 | 		ex.open(s.c_str());
2581 | 		extdContigs.clear();
2582 | 		extdTags.clear();
2583 | 		if(ex.is_open())
2584 | 		{
2585 | 			while(ex.good())
2586 | 			{
2587 | 				getline(ex, buf);
2588 | 				if(buf[0] == 0) break;
2589 | 
2590 | 				if(buf[0] == '>')
2591 | 				{
2592 | 					extdContigs.push_back(contig);
2593 | 					extdTags.push_back(0);
2594 | 				}
2595 | 				else
2596 | 					for(j = 0; j < buf.size(); j ++)
2597 | 						extdContigs[extdContigs.size() - 1].push_back(buf[j]);
2598 | 			}
2599 | 		}
2600 | 		else
2601 | 		{
2602 | 			cout << "CANNOT OPEN FILE!" << endl;
2603 | 			return;
2604 | 		}
2605 | 
2606 | 		s = "tmp/_initial_contigs." + itoa(i) + ".fa";
2607 | 		in.clear();
2608 | 	        in.open(s.c_str());
2609 | 		initContigs.clear();
2610 | 		initTags.clear();
2611 | 	        if(in.is_open())
2612 | 	        {
2613 | 	                while(in.good())
2614 | 	                {
2615 | 	                        getline(in, buf);
2616 | 	                        if(buf[0] == 0) break;
2617 | 
2618 | 	                        if(buf[0] == '>')
2619 | 	                        {
2620 | 	                                initContigs.push_back(contig);
2621 | 					initNums.push_back(atoi(buf.substr(1, buf.size()).c_str()));
2622 | 	                                initTags.push_back(0);
2623 | 	                        }
2624 | 	                        else
2625 | 	                                for(j = 0; j < buf.size(); j ++)
2626 | 	                                        initContigs[initContigs.size() - 1].push_back(buf[j]);
2627 | 	                }
2628 | 	        }
2629 | 	        else
2630 | 	        {
2631 | 	                cout << "CANNOT OPEN FILE!" << endl;
2632 | 	                return;
2633 | 	        }
2634 | 
2635 | 		s = "tmp/_short_initial_contigs_extended_contigs." + itoa(i) + ".psl";
2636 | 		ps.clear();
2637 | 		ps.open(s.c_str());
2638 | 		if(ps.is_open())
2639 | 		{
2640 | 			while(ps.good())
2641 | 			{
2642 | 				getline(ps, buf);
2643 | 				if(buf[0] == 0) break;
2644 | 
2645 | 				parseBLAT(buf, targetID, targetStart, targetEnd, targetGap, sourceID, sourceStart, sourceEnd, sourceGap, sourceSize, seg, fr, targetSize);
2646 | 				if((double)(sourceEnd - sourceStart - sourceGap) / sourceSize >= 0.95 && (double)(targetEnd - targetStart - targetGap) / (double)(targetEnd - targetStart) >= 0.95 && targetSize > sourceSize + 100)
2647 | 				{
2648 | 					initTags[sourceID] = 1;
2649 | 					extdTags[targetID] = 1;
2650 | 				}
2651 | 			}
2652 | 		}
2653 | 		else
2654 | 		{
2655 | 			cout << "CANNOT OPEN FILE!" << endl;
2656 | 			return;
2657 | 		}
2658 | 
2659 | 		s = "tmp/_post_extended_contigs." + itoa(i) + ".fa";
2660 | 		e.clear();
2661 | 		e.open(s.c_str());
2662 | 		if(e.is_open())
2663 | 		{
2664 | 			for(j = 0; j < extdTags.size(); j ++)
2665 | 				if(extdTags[j] == 1)
2666 | 				{
2667 | 					e << ">" << seqID ++ << endl;
2668 | 					for(k = 0; k < extdContigs[j].size(); k ++)
2669 | 					{
2670 | 						e << extdContigs[j][k];
2671 | 						if(k == extdContigs[j].size() - 1 || (k + 1) % 60 == 0)
2672 | 							e << endl;
2673 | 					}
2674 | 				}
2675 | 		}
2676 | 		else
2677 | 		{
2678 | 			cout << "CANNOT OPEN FILE!" << endl;
2679 | 			return;
2680 | 		}
2681 | 
2682 | 		s = "tmp/_post_initial_contigs." + itoa(i) + ".fa";
2683 | 		ini.clear();
2684 | 		ini.open(s.c_str());
2685 | 	        if(ini.is_open())
2686 | 	        {
2687 | 	                for(j = 0; j < initTags.size(); j ++)
2688 | 	                        if(initTags[j] == 1)
2689 | 	                        {
2690 | 	                                ini << ">" << initNums[j] << endl;
2691 | 	                                for(k = 0; k < initContigs[j].size(); k ++)
2692 | 	                                {
2693 | 	                                        ini << initContigs[j][k];
2694 | 	                                        if(k == initContigs[j].size() - 1 || (k + 1) % 60 == 0)
2695 | 	                                                ini << endl;
2696 | 	                                }
2697 | 	                        }
2698 | 	        }
2699 | 	        else
2700 | 	        {
2701 | 	                cout << "CANNOT OPEN FILE!" << endl;
2702 | 	                return;
2703 | 	        }
2704 | 
2705 | 		ex.close();
2706 | 		ps.close();
2707 | 		e.close();
2708 | 		ini.close();
2709 | 	}
2710 | }
2711 | 
2712 | void ref2(ofstream & e, ofstream & r)
2713 | {
2714 |         int numChromosomes, i, j, k, seqID;
2715 |         string s;
2716 |         vector<vector<char> > initContigs, extdContigs;
2717 |         vector<char> contig;
2718 |         string buf;
2719 |         unsigned int sourceID, targetID, targetStart, targetEnd, targetGap, sourceStart, sourceEnd, sourceGap, sourceSize, targetSize, fr;
2720 |         vector<int> initTags, extdTags;
2721 |         ifstream in, ex, ps;
2722 |         vector<Segment> seg;
2723 | 
2724 |         numChromosomes = readLog();
2725 |         for(i = 0; i < numChromosomes; i ++)
2726 |         {
2727 |                 s = "pblat tmp/_post_extended_contigs." + itoa(i) + ".fa tmp/_post_initial_contigs." + itoa(i) + ".fa -noHead tmp/_initial_contigs_extended_contigs." + itoa(i) + ".psl -threads=8 > blat_doc.txt 2> blat_doc.txt";
2728 |                 if(system(s.c_str()) != 0)
2729 | 		{
2730 | 			s = "blat tmp/_post_extended_contigs." + itoa(i) + ".fa tmp/_post_initial_contigs." + itoa(i) + ".fa -noHead tmp/_initial_contigs_extended_contigs." + itoa(i) + ".psl > blat_doc.txt 2> blat_doc.txt";
2731 | 			if(system(s.c_str()) != 0) {cout << "BLAT CALL FAILED!" << endl; exit(-1);}
2732 | 		}
2733 |         }
2734 | 
2735 |         in.open("tmp/_contigs.fa");
2736 |         if(in.is_open())
2737 |         {
2738 |                 while(in.good())
2739 |                 {
2740 |                         getline(in, buf);
2741 |                         if(buf[0] == 0) break;
2742 | 
2743 |                         if(buf[0] == '>')
2744 |                         {
2745 |                                 initContigs.push_back(contig);
2746 |                                 initTags.push_back(0);
2747 |                         }
2748 |                         else
2749 |                                 for(i = 0; i < buf.size(); i++)
2750 |                                         initContigs[initContigs.size() - 1].push_back(buf[i]);
2751 |                 }
2752 |         }
2753 |         else
2754 |         {
2755 |                 cout << "CANNOT OPEN FILE!" << endl;
2756 |                 return;
2757 |         }
2758 | 
2759 |         seqID = 0;
2760 |         for(i = 0; i < numChromosomes; i ++)
2761 |         {
2762 |                 s = "tmp/_post_extended_contigs." + itoa(i) + ".fa";
2763 | 		ex.clear();
2764 |                 ex.open(s.c_str());
2765 |                 extdContigs.clear();
2766 |                 extdTags.clear();
2767 |                 if(ex.is_open())
2768 |                 {
2769 |                         while(ex.good())
2770 |                         {
2771 |                                 getline(ex, buf);
2772 |                                 if(buf[0] == 0) break;
2773 | 
2774 |                                 if(buf[0] == '>')
2775 |                                 {
2776 |                                         extdContigs.push_back(contig);
2777 |                                         extdTags.push_back(0);
2778 |                                 }
2779 |                                 else
2780 |                                         for(j = 0; j < buf.size(); j ++)
2781 |                                                 extdContigs[extdContigs.size() - 1].push_back(buf[j]);
2782 |                         }
2783 |                 }
2784 |                 else
2785 |                 {
2786 |                         cout << "CANNOT OPEN FILE!" << endl;
2787 |                         return;
2788 |                 }
2789 | 
2790 |                 s = "tmp/_post_initial_contigs_post_extended_contigs." + itoa(i) + ".psl";
2791 | 		ps.clear();
2792 |                 ps.open(s.c_str());
2793 |                 if(ps.is_open())
2794 |                 {
2795 |                         while(ps.good())
2796 |                         {
2797 |                                 getline(ps, buf);
2798 |                                 if(buf[0] == 0) break;
2799 | 
2800 |                                 parseBLAT(buf, targetID, targetStart, targetEnd, targetGap, sourceID, sourceStart, sourceEnd, sourceGap, sourceSize, seg, fr, targetSize);
2801 |                                 if((double)(sourceEnd - sourceStart - sourceGap) / sourceSize >= 0.95 && (double)(targetEnd - targetStart - targetGap) / (double)(targetEnd - targetStart) >= 0.95 && targetSize > sourceSize + 100)
2802 |                                 {
2803 |                                         initTags[sourceID] = 1;
2804 |                                         extdTags[targetID] = 1;
2805 |                                 }
2806 |                         }
2807 |                 }
2808 |                 else
2809 |                 {
2810 |                         cout << "CANNOT OPEN FILE!" << endl;
2811 |                         return;
2812 |                 }
2813 | 
2814 |                 if(e.is_open())
2815 |                 {
2816 |                         for(j = 0; j < extdTags.size(); j ++)
2817 |                                 if(extdTags[j] == 1)
2818 |                                 {
2819 |                                         e << ">" << i << ": " << seqID ++ << endl;
2820 |                                         for(k = 0; k < extdContigs[j].size(); k ++)
2821 |                                         {
2822 |                                                 e << extdContigs[j][k];
2823 |                                                 if(k == extdContigs[j].size() - 1 || (k + 1) % 60 == 0)
2824 |                                                         e << endl;
2825 |                                         }
2826 |                                 }
2827 |                 }
2828 |                 else
2829 |                 {
2830 |                         cout << "CANNOT OPEN FILE!" << endl;
2831 |                         return;
2832 |                 }
2833 |                 ex.close();
2834 |                 ps.close();
2835 |         }
2836 | 
2837 |         if(r.is_open())
2838 |         {
2839 |                 for(i = 0; i < initTags.size(); i ++)
2840 |                         if(initTags[i] == 0)
2841 |                         {
2842 |                                 r << ">" << i << endl;
2843 |                                 for(j = 0; j < initContigs[i].size(); j ++)
2844 |                                 {
2845 |                                         r << initContigs[i][j];
2846 |                                         if(j == initContigs[i].size() - 1 || (j + 1) % 60 == 0)
2847 |                                                 r << endl;
2848 |                                 }
2849 |                         }
2850 |         }
2851 |         else
2852 |         {
2853 |                 cout << "CANNOT OPEN FILE!" << endl;
2854 |                 return;
2855 |         }
2856 | }
2857 | 
2858 | void refinement(ofstream & e, ofstream & r)
2859 | {
2860 | 	ref1();
2861 | 	ref2(e, r);
2862 | }
2863 | 
2864 | void refinement(ofstream & e, ofstream & r, int fastMap, int uniqueExtension, int numChromosomes)
2865 | {
2866 |         int i, j, k, seqID;
2867 |         string s, s0;
2868 |         vector<vector<char> > initContigs, extdContigs;
2869 |         vector<char> contig;
2870 |         string buf;
2871 |         unsigned int sourceID, targetID, targetStart, targetEnd, targetGap, sourceStart, sourceEnd, sourceGap, sourceSize, targetSize, fr, realSourceSize;
2872 |         vector<int> initTags, extdTags;
2873 |         ifstream in, ex, ps, cf;
2874 |         vector<Segment> seg;
2875 | 	vector<int> initNums;
2876 | 	int ID, IDBak = -1, targetIDBak;
2877 | 	vector<vector<int> > extdInitMap;
2878 | 	vector<int> eim;
2879 | 
2880 | #ifdef TEST
2881 | 	ofstream ini, ext;
2882 | 	ini.open("in.fa");
2883 | 	ext.open("ex.fa");
2884 | #endif
2885 | 
2886 | ///////////////////////////////////////////////////////////////////for easy alignment/////////////////////////////////////////////////////////////////////////////////////
2887 | 
2888 |         ifstream tmpIn;
2889 |         ofstream tmpOut;
2890 |         vector<vector<char> > tmpContigs;
2891 |         for(i = 0; i < numChromosomes; i ++)
2892 |         {
2893 |                 tmpIn.clear();
2894 |                 tmpOut.clear();
2895 |                 tmpContigs.clear();
2896 | 		initNums.clear();
2897 |                 seqID = 0;
2898 | 
2899 |                 s = "tmp/_initial_contigs." + itoa(i) + ".fa";
2900 |                 tmpIn.open(s.c_str());
2901 |                 s = "tmp/_short_initial_contigs." + itoa(i) + ".fa";
2902 |                 tmpOut.open(s.c_str());
2903 |                 if(tmpIn.is_open())
2904 |                 {
2905 |                         while(tmpIn.good())
2906 |                         {
2907 |                                 getline(tmpIn, buf);
2908 |                                 if(buf[0] == 0) break;
2909 | 
2910 |                                 if(buf[0] == '>')
2911 | 				{
2912 |                                         tmpContigs.push_back(contig);
2913 | 					initNums.push_back(atoi(buf.substr(1, buf.size()).c_str()));
2914 | 				}
2915 |                                 else
2916 |                                         for(j = 0; j < buf.size(); j ++)
2917 |                                                 tmpContigs[tmpContigs.size() - 1].push_back(buf[j]);
2918 |                         }
2919 |                 }
2920 |                 else
2921 |                 {
2922 |                         cout << "CANNOT OPEN FILE!" << endl;
2923 |                         exit(-1);
2924 |                 }
2925 | 
2926 |                 for(j = 0; j < tmpContigs.size(); j ++)
2927 |                 {
2928 | 			
2929 |                         if(tmpContigs[j].size() > SMALL_CHUNK)
2930 |                         {
2931 | 				tmpOut << ">" << initNums[j] << "." << tmpContigs[j].size()  << endl;// preserve id and size; parse for short contigs in a different way
2932 |                                 for(k = 0; k < SMALL_CHUNK; k ++)
2933 |                                 {
2934 |                                         tmpOut << tmpContigs[j][k];
2935 |                                         if((k + 1) % 60 == 0 || k == SMALL_CHUNK - 1)
2936 |                                                 tmpOut << endl;
2937 |                                 }
2938 |                         }
2939 |                         else
2940 |                         {
2941 | 				tmpOut << ">" << initNums[j] << endl;
2942 |                                 for(k = 0; k < tmpContigs[j].size(); k ++)
2943 |                                 {
2944 |                                         tmpOut << tmpContigs[j][k];
2945 |                                         if((k + 1) % 60 == 0 || k == tmpContigs[j].size() - 1)
2946 |                                                 tmpOut << endl;
2947 |                                 }
2948 |                         }
2949 |                 }
2950 | 
2951 |                 tmpIn.close();
2952 |                 tmpOut.close();
2953 |         }
2954 | 
2955 | //////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
2956 | 
2957 | 	if(fastMap == 1)
2958 | 	{
2959 | //NUCMER
2960 | 		for(i = 0; i < numChromosomes; i ++)
2961 | 		{
2962 | 			s = "nucmer tmp/_extended_contigs." + itoa(i) + ".fa tmp/_short_initial_contigs." + itoa(i) + ".fa -p tmp/_short_initial_contigs_extended_contigs." + itoa(i) + " > nucmer_doc.txt 2> nucmer_doc.txt";
2963 | 			if(system(s.c_str()) != 0) 
2964 | 			{
2965 | 				s = "touch tmp/_short_initial_contigs_extended_contigs." + itoa(i) + ".delta";
2966 | 				system(s.c_str());
2967 | 			}
2968 | 			s = "tmp/_short_initial_contigs_extended_contigs." + itoa(i) + ".delta"; 
2969 | 			delta2psl(s);
2970 | 		}
2971 | 	}
2972 | 	else
2973 | 	{
2974 | 	        for(i = 0; i < numChromosomes; i ++)
2975 | 	        {
2976 | 	                s = "pblat tmp/_extended_contigs." + itoa(i) + ".fa tmp/_short_initial_contigs." + itoa(i) + ".fa -noHead tmp/_short_initial_contigs_extended_contigs." + itoa(i) + ".psl -fastMap -threads=8 > blat_doc.txt 2> blat_doc.txt";
2977 | 	                if(system(s.c_str()) != 0)
2978 | 			{
2979 | 				s = "blat tmp/_extended_contigs." + itoa(i) + ".fa tmp/_short_initial_contigs." + itoa(i) + ".fa -noHead tmp/_short_initial_contigs_extended_contigs." + itoa(i) + ".psl -fastMap > blat_doc.txt 2> blat_doc.txt";
2980 | 				if(system(s.c_str()) != 0) {cout << "BLAT CALL FAILED!" << endl; exit(-1);}
2981 | 			}
2982 | 	        }
2983 | 	}
2984 | 
2985 |         in.open("tmp/_contigs.fa");
2986 |         if(in.is_open())
2987 |         {
2988 |                 while(in.good())
2989 |                 {
2990 |                         getline(in, buf);
2991 |                         if(buf[0] == 0) break;
2992 | 
2993 |                         if(buf[0] == '>')
2994 |                         {
2995 | 				for(i = 0; i < buf.size(); i ++)
2996 | 					if(buf[i] == '.') break;
2997 | 				ID = atoi(buf.substr(i + 1, buf.size()).c_str());
2998 | 				if(ID != IDBak)
2999 | 				{
3000 |                                 	initContigs.push_back(contig);
3001 |                                 	initTags.push_back(0);
3002 | 					IDBak = ID;
3003 | 				}
3004 |                         }
3005 |                         else
3006 |                                 for(i = 0; i < buf.size(); i++)
3007 |                                         initContigs[initContigs.size() - 1].push_back(buf[i]);
3008 |                 }
3009 |         }
3010 |         else
3011 |         {
3012 |                 cout << "CANNOT OPEN FILE!" << endl;
3013 |                 return;
3014 |         }
3015 | 
3016 | 	int line;
3017 | 	seqID = 0;
3018 |         for(i = 0; i < numChromosomes; i ++)
3019 |         {
3020 |                 s = "tmp/_extended_contigs." + itoa(i) + ".fa";
3021 |                 ex.open(s.c_str());
3022 |                 extdContigs.clear();
3023 |                 extdTags.clear();
3024 |                 if(ex.is_open())
3025 |                 {
3026 |                         while(ex.good())
3027 |                         {
3028 |                                 getline(ex, buf);
3029 |                                 if(buf[0] == 0) break;
3030 | 
3031 |                                 if(buf[0] == '>')
3032 |                                 {
3033 |                                         extdContigs.push_back(contig);
3034 |                                         extdTags.push_back(0);
3035 | 					extdInitMap.push_back(eim);
3036 |                                 }
3037 |                                 else
3038 |                                         for(j = 0; j < buf.size(); j ++)
3039 |                                                 extdContigs[extdContigs.size() - 1].push_back(buf[j]);
3040 |                         }
3041 |                 }
3042 |                 else
3043 |                 {
3044 |                         cout << "CANNOT OPEN FILE!" << endl;
3045 |                         return;
3046 |                 }
3047 | 
3048 | 		targetIDBak = -1;
3049 |                 s = "tmp/_short_initial_contigs_extended_contigs." + itoa(i) + ".psl";
3050 |                 ps.open(s.c_str());
3051 |                 if(ps.is_open())
3052 |                 {
3053 |                         while(ps.good())
3054 |                         {
3055 |                                 getline(ps, buf);
3056 |                                 if(buf[0] == 0) break;
3057 | 
3058 |                                 realSourceSize = parseBLAT(buf, targetID, targetStart, targetEnd, targetGap, sourceID, sourceStart, sourceEnd, sourceGap, sourceSize, seg, fr, targetSize);
3059 |                                 if((double)(sourceEnd - sourceStart - sourceGap) / sourceSize >= 0.8 && (double)(targetEnd - targetStart - targetGap) / (double)(targetEnd - targetStart) >= 0.8 && targetSize > realSourceSize + 100 && realSourceSize > targetSize / 100)
3060 |                                 {
3061 |                                         if(uniqueExtension == 1)
3062 |                                         {
3063 |                                                 if(initTags[sourceID] > 0 && targetIDBak != -1)
3064 |                                                 {
3065 |                                                         if(extdTags[targetIDBak] < extdTags[targetID])
3066 |                                                         {
3067 |                                                                 extdTags[targetIDBak] = 0;
3068 | 								extdInitMap[targetIDBak].pop_back();
3069 |                                                                 extdTags[targetID] = targetSize;
3070 |                                                                 initTags[sourceID] = 1;
3071 | 								extdInitMap[targetID].push_back(sourceID);
3072 |                                                         }
3073 |                                                 }
3074 |                                                 else
3075 |                                                 {
3076 |                                                         extdTags[targetID] = targetSize;
3077 |                                                         initTags[sourceID] = 1;
3078 | 							extdInitMap[targetID].push_back(sourceID);
3079 |                                                 }
3080 |                                                 targetIDBak = targetID;
3081 |                                         }
3082 |                                         else
3083 |                                         {
3084 |                                                 extdTags[targetID] = 1;
3085 |                                                 initTags[sourceID] = 1;
3086 | 						extdInitMap[targetID].push_back(sourceID);
3087 |                                         }
3088 |                                 }
3089 |                         }
3090 |                 }
3091 |                 else
3092 |                 {
3093 |                         cout << "CANNOT OPEN FILE!" << endl;
3094 |                         return;
3095 |                 }
3096 | 
3097 |                 if(e.is_open())
3098 |                 {
3099 |                         for(j = 0; j < extdTags.size(); j ++)
3100 |                                 if(extdTags[j] > 0)
3101 |                                 {
3102 | 					e << ">" << "AlignGraph" << seqID << " @ " << genomeIds[i] << " : ";
3103 | 					for(k = 0; k < extdInitMap[j].size(); k ++)
3104 | 						e << contigIds[extdInitMap[j][k]] << " ; ";
3105 | 					e << endl;
3106 | #ifdef TEST
3107 | 					ext << ">" << i << ": " << seqID << endl;
3108 | #endif
3109 | 					seqID ++;
3110 |                                         for(k = 0; k < extdContigs[j].size(); k ++)
3111 |                                         {
3112 |                                                 e << extdContigs[j][k];
3113 | #ifdef TEST
3114 | 						ext << extdContigs[j][k];
3115 | #endif
3116 |                                                 if(k == extdContigs[j].size() - 1 || (k + 1) % 60 == 0)
3117 | 						{
3118 |                                                         e << endl;
3119 | #ifdef TEST
3120 | 							ext << endl;
3121 | #endif
3122 | 						}
3123 |                                         }
3124 |                                 }
3125 |                 }
3126 |                 else
3127 |                 {
3128 |                         cout << "CANNOT OPEN FILE!" << endl;
3129 |                         return;
3130 |                 }
3131 |                 ex.close();
3132 |                 ps.close();
3133 |         }
3134 | 
3135 |         if(r.is_open())
3136 |         {
3137 |                 for(i = 0; i < initTags.size(); i ++)
3138 |                         if(initTags[i] == 0)
3139 |                         {
3140 | 				r << ">" << contigIds[i] << endl;
3141 |                                 for(j = 0; j < initContigs[i].size(); j ++)
3142 |                                 {
3143 |                                         r << initContigs[i][j];
3144 |                                         if(j == initContigs[i].size() - 1 || (j + 1) % 60 == 0)
3145 |                                                 r << endl;
3146 |                                 }
3147 |                         }
3148 | 
3149 | 		cf.open("tmp/_chaff.fa");
3150 | 		if(cf.is_open())
3151 | 		{
3152 | 			while(cf.good())
3153 | 			{
3154 | 				getline(cf, buf);
3155 | 				if(buf[0] == 0) break;
3156 | 
3157 | 				for(i = 0; i < buf.size(); i ++)
3158 | 					r << buf[i];
3159 | 				r << endl;
3160 | 			}
3161 | 		}
3162 | 		else
3163 | 		{
3164 | 			cout << "CANNOT OPEN FILE!" << endl;
3165 | 			return;
3166 | 		}
3167 |         }
3168 |         else
3169 |         {
3170 |                 cout << "CANNOT OPEN FILE!" << endl;
3171 |                 return;
3172 |         }
3173 | 
3174 | #ifdef TEST
3175 |         if(ini.is_open())
3176 |         {
3177 |                 for(i = 0; i < initTags.size(); i ++)
3178 |                         if(initTags[i] == 1)
3179 |                         {
3180 |                                 ini << ">" << i << endl;
3181 |                                 for(j = 0; j < initContigs[i].size(); j ++)
3182 |                                 {
3183 |                                         ini << initContigs[i][j];
3184 |                                         if(j == initContigs[i].size() - 1 || (j + 1) % 60 == 0)
3185 |                                                 ini << endl;
3186 |                                 }
3187 |                         }
3188 |         }
3189 |         else
3190 |         {
3191 |                 cout << "CANNOT OPEN FILE!" << endl;
3192 |                 return;
3193 |         }
3194 | #endif
3195 | }
3196 | 
3197 | int maxReadLength(ifstream & r)
3198 | {
3199 | 	int max = 0, len = 0;
3200 | 	string buf;
3201 | 
3202 | 	if(r.is_open())
3203 | 	{
3204 | 		while(r.good())
3205 | 		{
3206 | 			getline(r, buf);
3207 | 			if(buf[0] == 0) break;
3208 | 			if(buf[0] == '>') 
3209 | 			{
3210 | 				if(len > max) 
3211 | 					max = len; 
3212 | 				len = 0; 
3213 | 				continue;
3214 | 			}
3215 | 			len = len + buf.size();
3216 | 		}
3217 | 		if(len > max) max = len;
3218 | 	}
3219 | 	else
3220 | 	{
3221 | 		cout << "CANNOT OPEN FILE!" << endl;
3222 | 		exit(-1);
3223 | 	}
3224 | 
3225 | 	return max;
3226 | }
3227 | 
3228 | void formalizeInput(ifstream & in, string file)
3229 | {
3230 |         string buf;
3231 |         int i, cp, cpp, total, sp;
3232 |         unsigned long seqID = 0, count = 0, realSeqID = 0;
3233 | 	vector<vector<char> > contigs;
3234 | 	vector<char> contig;
3235 |         ofstream out, addOut;
3236 | 	string id;
3237 | 	vector<string> cIds;
3238 | 
3239 | 	in.clear();
3240 | 	in.seekg(0);
3241 |         out.open(file.c_str());
3242 | 	if(file == "tmp/_contigs.fa") addOut.open("tmp/_chaff.fa");
3243 | 
3244 | 	if(file == "tmp/_contigs.fa" || file == "tmp/_extended_contigs.fa" || file == "tmp/_remaining_contigs.fa")
3245 | 	{
3246 | 	        if(in.is_open())
3247 | 	        {
3248 | 	                while(in.good())
3249 | 	                {
3250 | 	                        getline(in, buf);
3251 | 	                        if(buf[0] == 0)
3252 | 	                                break;
3253 | 	                        if(buf[0] == '>')
3254 | 	                        {
3255 | 					contigs.push_back(contig);
3256 | 					cIds.push_back(buf.substr(1, buf.size()));
3257 | 	                        }
3258 | 	                        else
3259 | 					for(i = 0; i < buf.size(); i ++)
3260 | 						contigs[contigs.size() - 1].push_back(buf[i]);
3261 | 	                }
3262 | 
3263 | 			for(cp = 0; cp < contigs.size(); cp ++)
3264 | 			{
3265 | 				if(contigs[cp].size() > 200)
3266 | 				{
3267 | 					if(contigs[cp].size() < LARGE_CHUNK)
3268 | 					{
3269 | 						out << ">" << seqID ++ << "." << realSeqID << endl;
3270 | 						for(cpp = 0; cpp < contigs[cp].size(); cpp ++)
3271 | 						{
3272 | 							out << contigs[cp][cpp];
3273 | 							if((cpp + 1) % 60 == 0 || cpp == contigs[cp].size() - 1)
3274 | 								out << endl;
3275 | 						}
3276 | 					}
3277 | 					else
3278 | 					{
3279 | 						out << ">" << seqID ++ << "." << realSeqID << endl;
3280 | 						for(cpp = 0, total = 0; cpp < contigs[cp].size(); cpp ++)
3281 | 						{
3282 | 							out << contigs[cp][cpp];
3283 | 							if((cpp + 1) % LARGE_CHUNK == 0 && cpp < contigs[cp].size() - 1 - 60)
3284 | 							{
3285 | 								total = total + LARGE_CHUNK;
3286 | 								out << endl;
3287 | 								out << ">" << seqID ++ << "." << realSeqID << endl;
3288 | 								continue;
3289 | 							}
3290 | 							if((cpp + 1 - total) % 60 == 0 || cpp == contigs[cp].size() - 1)
3291 | 								out << endl;
3292 | 						}
3293 | 					}
3294 | 					realSeqID ++;
3295 | 				}
3296 | 				else
3297 | 				{
3298 | 					addOut << ">" << cIds[cp] << endl;
3299 | 					for(cpp = 0; cpp < contigs[cp].size(); cpp ++)
3300 | 					{
3301 | 						addOut << contigs[cp][cpp];
3302 | 						if((cpp + 1) % 60 == 0 || cpp == contigs[cp].size() - 1)
3303 | 							addOut << endl;
3304 | 					}
3305 | 					cIds[cp] = "\0";
3306 | 				}
3307 | 			}
3308 | 
3309 | 			contigIds.clear();
3310 | 			for(sp = 0; sp < cIds.size(); sp ++)
3311 | 				if(cIds[sp] != "\0")
3312 | 					contigIds.push_back(cIds[sp]);
3313 | 	        }
3314 | 	        else
3315 | 	        {
3316 | 	                cout << "CANNOT OPEN FILE!" << endl;
3317 | 	                exit(-1);
3318 | 	        }
3319 | 	}
3320 | 	else
3321 | 	{
3322 | 		if(in.is_open())
3323 | 		{
3324 | 			while(in.good())
3325 | 			{
3326 | 				getline(in, buf);
3327 | 				if(buf[0] == 0)
3328 | 					break;
3329 | 				if(buf[0] == '>')
3330 | 					out << ">" << seqID ++ << endl;
3331 | 				else
3332 | 				{
3333 | 					for(i = 0; i < buf.size(); i ++)
3334 | 						out << buf[i];
3335 | 					out << endl;
3336 | 				}
3337 | 			}
3338 | 		}
3339 | 		else
3340 | 		{
3341 | 			cout << "CANNOT OPEN FILE!" << endl;
3342 | 			exit(-1);
3343 | 		}
3344 | 	}
3345 | }
3346 | 
3347 | int formalizeGenome(ifstream & in, int p)
3348 | {
3349 |         string buf;
3350 |         int i, chromosomeID, gp, cp, q;
3351 |         ofstream out, out0;
3352 | 	string s;
3353 | 	vector<vector<char> > genome;
3354 | 	vector<char> g;
3355 | 	vector<string> id;
3356 | 
3357 |         if(in.is_open())
3358 |         {
3359 |                 while(in.good())
3360 |                 {
3361 |                         getline(in, buf);
3362 |                         if(buf[0] == 0)
3363 |                                 break;
3364 |                         if(buf[0] == '>')
3365 |                         {
3366 | 				genome.push_back(g);
3367 | 				genomeIds.push_back(buf.substr(1, buf.size()));
3368 |                         }
3369 |                         else
3370 | 			{
3371 |                                 for(i = 0; i < buf.size(); i ++)
3372 |                                         genome[genome.size() - 1].push_back(buf[i]);
3373 | 			}
3374 |                 }
3375 |         }
3376 |         else
3377 |         {
3378 |                 cout << "CANNOT OPEN FILE!" << endl;
3379 |                 exit(-1);
3380 |         }
3381 | 
3382 | 	chromosomeID = 0;
3383 | 	out0.open("tmp/_genome.fa");
3384 | 	for(gp = 0; gp < genome.size(); gp ++)
3385 | 	{
3386 | 		s = "tmp/_genome." + itoa(chromosomeID) + ".fa";
3387 | 		out.open(s.c_str());
3388 | 		out << ">0" << endl;
3389 | 		out0 << ">" << chromosomeID << endl;
3390 | 		q = 1;
3391 | 		for(cp = 0; cp < genome[gp].size(); cp ++)
3392 | 		{
3393 | 			out << genome[gp][cp];
3394 | 			out0 << genome[gp][cp];
3395 | 			if((cp + 1) % 60 == 0 || cp == genome[gp].size() - 1 || ((cp + 1) % (genome[gp].size() / p) == 0 && q < p))
3396 | 			{
3397 | 				out << endl;
3398 | 				out0 << endl;
3399 | 			}
3400 | 			if(cp != genome[gp].size() - 1 && ((cp + 1) % (genome[gp].size() / p) == 0 && q < p))
3401 | 			{
3402 | 				out.close();
3403 | 				chromosomeID ++;
3404 | 				q ++;
3405 | 				s = "tmp/_genome." + itoa(chromosomeID) + ".fa";
3406 | 				out.open(s.c_str());
3407 | 				out << ">0" << endl;
3408 | 				out0 << ">" << chromosomeID << endl; 
3409 | 			}
3410 | 		}
3411 | 		out.close();
3412 | 		chromosomeID ++;
3413 | 	}
3414 | 	genome.clear();
3415 | 	out.close();
3416 | 	out0.close();
3417 | 	return chromosomeID;
3418 | }
3419 | 
3420 | int formalizeInput(ifstream & in1, ifstream & in2, string file, string file1, string file2)
3421 | {
3422 |         string buf1, buf2;
3423 |         int i, size;
3424 |         unsigned long seqID = 0;
3425 |         ofstream out, out1, out2;
3426 | 	vector<char> read1, read2;
3427 | 
3428 | 	in1.clear();
3429 | 	in2.clear();
3430 | 	in1.seekg(0);
3431 | 	in2.seekg(0);
3432 |         out.open(file.c_str());
3433 | 	out1.open(file1.c_str());
3434 | 	out2.open(file2.c_str());
3435 | 
3436 |         if(in1.is_open() && in2.is_open())
3437 |         {
3438 |                 while(in1.good() && in2.good())
3439 |                 {
3440 |                         getline(in1, buf1);
3441 | 			getline(in2, buf2);
3442 |                         if(buf1[0] == 0 && buf2[0] == 0)
3443 |                                 break;
3444 | 			else if(buf1[0] == 0 && buf2[0] != 0 || buf1[0] != 0 && buf2[0] == 0)
3445 | 			{
3446 | 				cout << "INCONSISTENT PE FILES!" << endl;
3447 | 				exit(-1);
3448 | 			}
3449 | 
3450 |                         if(buf1[0] == '>' && buf2[0] == '>')
3451 |                         {
3452 | 				if(read1.size() != 0 && read2.size() != 0)
3453 | 				{
3454 | 					size = read1.size() < read2.size() ? read1.size() : read2.size();
3455 | 					out << ">" << seqID << endl;
3456 | 					for(i = 0; i < size; i ++)
3457 | 						out << read1[i];
3458 | 					out << endl;
3459 | 					out1 << ">" << seqID << endl;
3460 | 					for(i = 0; i < size; i ++)
3461 | 						out1 << read1[i];
3462 | 					out1 << endl;
3463 | 
3464 | 					out << ">" << seqID << endl;
3465 | 					for(i = 0; i < size; i ++)
3466 | 						out << read2[i];
3467 | 					out << endl;
3468 | 					out2 << ">" << seqID ++ << endl;
3469 | 					for(i = 0; i < size; i ++)
3470 | 						out2 << read2[i];
3471 | 					out2 << endl;
3472 | 				}
3473 | 
3474 | 				read1.clear();
3475 | 				read2.clear();
3476 |                         }
3477 |                         else if(buf1[0] != '>' && buf2[0] != '>')
3478 |                         {
3479 | 				for(i = 0; i < buf1.size(); i ++)
3480 | 					read1.push_back(buf1[i]);
3481 | 				for(i = 0; i < buf2.size(); i ++)
3482 | 					read2.push_back(buf2[i]);
3483 |                         }
3484 | 			else
3485 | 			{
3486 | 				cout << "INCONSISTENT PE FILES!" << endl;
3487 | 				exit(-1);
3488 | 			}
3489 | 		}
3490 | 		if(read1.size() != 0 && read2.size() != 0)
3491 |                 {
3492 | 			size = read1.size() < read2.size() ? read1.size() : read2.size();
3493 |                         out << ">" << seqID << endl;
3494 |                         for(i = 0; i < size; i ++)
3495 |                                 out << read1[i];
3496 |                         out << endl;
3497 | 			out1 << ">" << seqID << endl;
3498 | 			for(i = 0; i < size; i ++)
3499 | 				out1 << read1[i];
3500 | 			out1 << endl;
3501 | 
3502 |                         out << ">" << seqID << endl;
3503 |                         for(i = 0; i < size; i ++)
3504 |                                 out << read2[i];
3505 |                         out << endl;
3506 | 			out2 << ">" << seqID ++ << endl;
3507 | 			for(i = 0; i < size; i ++)
3508 | 				out2 << read2[i];
3509 | 			out2 << endl;
3510 |                 }
3511 |         }
3512 |         else
3513 |         {
3514 |                 cout << "CANNOT OPEN FILE!" << endl;
3515 |                 exit(-1);
3516 |         }
3517 | 	return seqID;
3518 | }
3519 | 
3520 | int parseBT(string buf)
3521 | {
3522 | 	int i, item = 0, j = 0;
3523 | 	char targetIDBuf[10] = {'\0'};
3524 | 
3525 | 	for(i = 0; i < buf.size(); i ++)
3526 | 	{
3527 | 		if(buf[i] == '	')
3528 | 		{
3529 | 			item ++;
3530 | 			continue;
3531 | 		}
3532 | 
3533 |                 if(item == 2)
3534 |                 {
3535 |                         if(buf[i] == '*') return -1;
3536 | 
3537 |                         targetIDBuf[j ++] = buf[i];
3538 |                 }
3539 | 
3540 | 		if(item == 3) break;
3541 | 	}
3542 | 	return atoi(targetIDBuf);
3543 | }
3544 | 
3545 | void distributeAlignments(int numChromosomes)
3546 | {
3547 | 	ifstream in;
3548 | 	ofstream out;
3549 | 	string buf, s;
3550 | 	int chromosomeID;
3551 | 
3552 | 	in.open("tmp/_reads_genome.bowtie");
3553 | 	for(chromosomeID = 0; chromosomeID < numChromosomes; chromosomeID ++)
3554 | 	{
3555 | 		s = "tmp/_reads_genome." + itoa(chromosomeID) + ".bowtie";
3556 | 		out.open(s.c_str());
3557 | 		if(in.is_open())
3558 | 		{
3559 | 			while(in.good())
3560 | 			{
3561 | 				getline(in, buf);
3562 | 				if(buf[0] == '@') continue;
3563 | 				if(buf[0] == 0) break;
3564 | 
3565 | 				if(parseBT(buf) == chromosomeID)
3566 | 					out << buf << endl;
3567 | 			}
3568 | 		}
3569 | 		else
3570 | 		{
3571 | 			cout << "CANNOT OPEN FILE!" << endl;
3572 | 			exit(-1);
3573 | 		}
3574 | 		in.clear();
3575 | 		in.seekg(0);
3576 | 		out.close();
3577 | 	}
3578 | 	in.close();
3579 | }
3580 | 
3581 | void * task0(void * arg)
3582 | {
3583 |         Insert ins;
3584 |         string command;
3585 |         stringstream distanceLowStr, distanceHighStr;
3586 | 	int chromosomeID;
3587 | 
3588 |         ins = *(Insert *) arg;
3589 |         distanceLowStr << ins.distanceLow;
3590 |         distanceHighStr << ins.distanceHigh;
3591 |         ins.numChromosomes;
3592 |         ins.fastMap;
3593 | 	ins.iterativeMap;
3594 | 
3595 | 	if(ins.iterativeMap == 1)
3596 | 	{
3597 | 		for(chromosomeID = 0; chromosomeID < ins.numChromosomes; chromosomeID ++)
3598 | 		{
3599 | 			command = "bowtie2-build -f tmp/_genome." + itoa(chromosomeID) + ".fa tmp/_genome." + itoa(chromosomeID) + " > bowtie_doc.txt 2> bowtie_doc.txt";
3600 | 			system(command.c_str());
3601 | 			command = "bowtie2 -f --no-mixed -k 5 -p 8 --local --mp 3,1 --rdg 2,1 --rfg 2,1 --score-min G,5,2 -I " + distanceLowStr.str() + " -X " + distanceHighStr.str() + " --no-discordant -x tmp/_genome." + itoa(chromosomeID) + " -1 tmp/_reads_1.fa -2 tmp/_reads_2.fa --reorder > tmp/_reads_genome." + itoa(chromosomeID) + ".bowtie 2> bowtie_doc.txt";
3602 | 			system(command.c_str());
3603 | 		}
3604 | 	}
3605 | 	else
3606 | 	{
3607 | 		command = "bowtie2-build -f tmp/_genome.fa tmp/_genome > bowtie_doc.txt 2> bowtie_doc.txt";
3608 | 		system(command.c_str());
3609 | 		command = "bowtie2 -f --no-mixed -k 5 -p 8 --local --mp 3,1 --rdg 2,1 --rfg 2,1 --score-min G,5,2 -I " + distanceLowStr.str() + " -X " + distanceHighStr.str() + " --no-discordant -x tmp/_genome -1 tmp/_reads_1.fa -2 tmp/_reads_2.fa --reorder > tmp/_reads_genome.bowtie 2> bowtie_doc.txt";
3610 | 		system(command.c_str());
3611 | 		distributeAlignments(ins.numChromosomes);
3612 | 	}
3613 | }
3614 | 
3615 | void * task1(void * arg)
3616 | {
3617 | 	Insert ins;
3618 | 	int chromosomeID;
3619 | 	string command, s;
3620 | 	stringstream distanceLowStr, distanceHighStr;
3621 | 
3622 | 	ins = *(Insert *) arg;
3623 | 	distanceLowStr << ins.distanceLow;
3624 | 	distanceHighStr << ins.distanceHigh;
3625 | 	ins.numChromosomes;
3626 | 	ins.fastMap;
3627 | 	ins.iterativeMap;
3628 | 
3629 | 	if(ins.fastMap == 1)
3630 | 	{
3631 | //NUCMER
3632 | 		for(chromosomeID = 0; chromosomeID < ins.numChromosomes; chromosomeID ++)
3633 | 		{
3634 | 			command = "nucmer tmp/_genome." + itoa(chromosomeID) + ".fa tmp/_contigs.fa -p tmp/_contigs_genome." + itoa(chromosomeID) + " > nucmer_doc.txt 2> nucmer_doc.txt";
3635 | 			if(system(command.c_str()) != 0) 
3636 | 			{
3637 | 				command = "touch tmp/_contigs_genome." + itoa(chromosomeID) + ".delta";
3638 | 				system(command.c_str());
3639 | 			}
3640 | 			command = "tmp/_contigs_genome." + itoa(chromosomeID) + ".delta";
3641 | 			delta2psl(command);
3642 | 		}
3643 | 	}
3644 | 	else
3645 | 	{
3646 | 		for(chromosomeID = 0; chromosomeID < ins.numChromosomes; chromosomeID ++)
3647 | 		{
3648 | 			command = "pblat tmp/_genome." + itoa(chromosomeID) + ".fa tmp/_contigs.fa -noHead tmp/_contigs_genome." + itoa(chromosomeID) + ".psl -fastMap -threads=8 > blat_doc.txt 2> blat_doc.txt";
3649 | 			if(system(command.c_str()) != 0)
3650 | 			{
3651 | 				command = "blat tmp/_genome." + itoa(chromosomeID) + ".fa tmp/_contigs.fa -noHead tmp/_contigs_genome." + itoa(chromosomeID) + ".psl -fastMap > blat_doc.txt 2> blat_doc.txt";
3652 | 				if(system(command.c_str()) != 0) {cout << "BLAT CALL FAILED!" << endl; exit(-1);}
3653 | 			}
3654 | 		}
3655 | 	}
3656 | }
3657 | 
3658 | void nonParallelMap(int distanceLow, int distanceHigh, int numChromosomes, int fastMap, int iterativeMap)
3659 | {
3660 |         string command, s;
3661 |         stringstream distanceLowStr, distanceHighStr;
3662 | 	int chromosomeID;
3663 | 
3664 | 	distanceLowStr << distanceLow;
3665 | 	distanceHighStr << distanceHigh;
3666 | 
3667 | 	if(iterativeMap == 1)
3668 | 	{
3669 | 		for(chromosomeID = 0; chromosomeID < numChromosomes; chromosomeID ++)
3670 | 		{
3671 | 			command = "bowtie2-build -f tmp/_genome." + itoa(chromosomeID) + ".fa tmp/_genome." + itoa(chromosomeID) + " > bowtie_doc.txt 2> bowtie_doc.txt";
3672 | 			system(command.c_str());
3673 | 			command = "bowtie2 -f --no-mixed -k 5 -p 8 --local --mp 3,1 --rdg 2,1 --rfg 2,1 --score-min G,5,2 -I " + distanceLowStr.str() + " -X " + distanceHighStr.str() + " --no-discordant -x tmp/_genome." + itoa(chromosomeID) + " -1 tmp/_reads_1.fa -2 tmp/_reads_2.fa --reorder > tmp/_reads_genome." + itoa(chromosomeID) + ".bowtie 2>> bowtie_doc.txt";// --very-sensitive-local
3674 | //			command = "bowtie2 -f --no-mixed -k 5 -p 8 --end-to-end --mp 3,1 --rdg 2,1 --rfg 2,1 --score-min L,-0.24,-0.24 -I " + distanceLowStr.str() + " -X " + distanceHighStr.str() + " --no-discordant -x tmp/_genome." + itoa(chromosomeID) + " -1 tmp/_reads_1.fa -2 tmp/_reads_2.fa --reorder > tmp/_reads_genome." + itoa(chromosomeID) + ".bowtie 2> bowtie_doc.txt";
3675 | 			system(command.c_str());
3676 | 		}
3677 | 	}
3678 | 	else
3679 | 	{
3680 | 	        command = "bowtie2-build -f tmp/_genome.fa tmp/_genome > bowtie_doc.txt 2> bowtie_doc.txt";
3681 | 		system(command.c_str());
3682 | 	        command = "bowtie2 -f --no-mixed -k 5 -p 8 --local --mp 3,1 --rdg 2,1 --rfg 2,1 --score-min G,5,2 -I " + distanceLowStr.str() + " -X " + distanceHighStr.str() + " --no-discordant -x tmp/_genome -1 tmp/_reads_1.fa -2 tmp/_reads_2.fa --reorder > tmp/_reads_genome.bowtie 2> bowtie_doc.txt";
3683 | 	        system(command.c_str());
3684 | 	        distributeAlignments(numChromosomes);
3685 | 	}
3686 | 
3687 | 	if(fastMap == 1)
3688 | 	{
3689 | //NUCMER
3690 | 		command = "nucmer tmp/_genome." + itoa(chromosomeID) + ".fa tmp/_contigs.fa -p tmp/_contigs_genome." + itoa(chromosomeID) + " > nucmer_doc.txt 2> nucmer_doc.txt";
3691 | 		if(system(command.c_str()) != 0) 
3692 | 		{
3693 | 			command = "touch tmp/_contigs_genome." + itoa(chromosomeID) + ".delta";
3694 | 			system(command.c_str());
3695 | 		}
3696 | 		command = "tmp/_contigs_genome." + itoa(chromosomeID) + ".delta";
3697 | 		delta2psl(command);
3698 | 	}
3699 | 	else
3700 | 	{
3701 | 		for(chromosomeID = 0; chromosomeID < numChromosomes; chromosomeID ++)
3702 | 		{
3703 | //	              command = "lastdb -c -uMAM8 tmp/_genome." + itoa(chromosomeID) + ".db tmp/_genome." + itoa(chromosomeID) + ".fa > last_doc.txt";
3704 | //	              system(command.c_str());
3705 | //	              command = "lastal -e34 -m100 tmp/_genome." + itoa(chromosomeID) + ".db tmp/_contigs.fa | last-split > tmp/_contigs_genome." + itoa(chromosomeID) + ".maf 2>> last_doc.txt";
3706 | //	              system(command.c_str());
3707 | //	              command = "maf-convert.py psl tmp/_contigs_genome." + itoa(chromosomeID) + ".maf > tmp/_contigs_genome." + itoa(chromosomeID) + ".psl 2>> last_doc.txt";
3708 | //	              system(command.c_str());
3709 | 
3710 | 			command = "pblat tmp/_genome." + itoa(chromosomeID) + ".fa tmp/_contigs.fa -noHead tmp/_contigs_genome." + itoa(chromosomeID) + ".psl -fastMap -threads=8 > blat_doc.txt 2> blat_doc.txt";// -minIdentity=50 -q=dnax -t=dnax
3711 | 			if(system(command.c_str()) != 0)
3712 | 			{
3713 | 				command = "blat tmp/_genome." + itoa(chromosomeID) + ".fa tmp/_contigs.fa -noHead tmp/_contigs_genome." + itoa(chromosomeID) + ".psl -fastMap > blat_doc.txt 2> blat_doc.txt";
3714 | 				if(system(command.c_str()) != 0) {cout << "BLAT CALL FAILED!" << endl; exit(-1);}
3715 | 			}
3716 | 		}
3717 | 	}
3718 | }
3719 | 
3720 | void parallelMap(int distanceLow, int distanceHigh, int numChromosomes, int fastMap, int iterativeMap)
3721 | {
3722 | 	pthread_t t0, t1;
3723 | 	Insert ins;
3724 | 
3725 | 	ins.distanceLow = distanceLow;
3726 | 	ins.distanceHigh = distanceHigh;
3727 | 	ins.numChromosomes = numChromosomes;
3728 | 	ins.fastMap = fastMap;
3729 | 	ins.iterativeMap = iterativeMap;
3730 | 	if(pthread_create(&t0, NULL, task0, &ins) != 0) {nonParallelMap(distanceLow, distanceHigh, numChromosomes, fastMap, iterativeMap); return;}
3731 | 	if(pthread_create(&t1, NULL, task1, &ins) != 0) {nonParallelMap(distanceLow, distanceHigh, numChromosomes, fastMap, iterativeMap); return;}
3732 | 	
3733 | 	if(pthread_join(t0, NULL) != 0) {perror("Thread join faied"); exit(EXIT_FAILURE);}
3734 | 	if(pthread_join(t1, NULL) != 0) {perror("Thread join faied"); exit(EXIT_FAILURE);}
3735 | }
3736 | 
3737 | void writeLog(int n)
3738 | {
3739 | 	ofstream out;
3740 | 
3741 | 	out.open("tmp/log.txt");
3742 | 	if(out.is_open())
3743 | 		out << n << endl;
3744 | 	else
3745 | 	{
3746 | 		cout << "CANNOT OPEN FILE!" << endl;
3747 | 		exit(-1);
3748 | 	}
3749 | }
3750 | 
3751 | void checkRatio(int numChromosomes)
3752 | {
3753 | 	int i, rp, chromosomeID, aligned;
3754 | 	unsigned targetID1, targetStart1, targetEnd1, targetGap1, sourceID1, sourceStart1, sourceEnd1, sourceGap1, sourceSize1, fr1, targetID2, targetStart2, targetEnd2, targetGap2, sourceID2, sourceStart2, sourceEnd2, sourceGap2, sourceSize2, fr2;
3755 |         vector<Segment> segs1, segs2;
3756 | 	ifstream r, ra;
3757 | 	string s, buf;
3758 | 	vector<int> reads;
3759 | 	double ratio;
3760 | 
3761 | 	r.open("tmp/_reads_1.fa");
3762 | 	if(r.is_open())
3763 | 	{
3764 | 		while(r.good())
3765 | 		{
3766 | 			getline(r, buf);
3767 | 			if(buf[0] == '>')
3768 | 				reads.push_back(0);
3769 | 		}
3770 | 	}
3771 | 	else
3772 | 	{
3773 | 		cout << "CANNOT OPEN FILE!" << endl;
3774 | 		exit(-1);
3775 | 	}
3776 | 
3777 | 	for(chromosomeID = 0; chromosomeID < numChromosomes; chromosomeID ++)
3778 | 	{
3779 | 		s = "tmp/_reads_genome." + itoa(chromosomeID) + ".bowtie";
3780 | 		ra.clear();
3781 | 		ra.seekg(0);
3782 | 		ra.open(s.c_str());
3783 | 		if(ra.is_open())
3784 | 		{
3785 | 			while(ra.good())
3786 | 			{
3787 | 				getline(ra, buf);
3788 | 				if(buf[0] == 0) break;
3789 | 				if(buf[0] == '@') continue;
3790 | 				parseBOWTIE(buf, targetID1, targetStart1, targetEnd1, targetGap1, sourceID1, sourceStart1, sourceEnd1, sourceGap1, sourceSize1, segs1, fr1);
3791 | 
3792 | 				getline(ra, buf);
3793 | 				if(buf[0] == 0)
3794 | 				{
3795 | 					cout << "BROKEN BOWTIE FILE" << endl;
3796 | 					exit(-1);
3797 | 				}
3798 | 				parseBOWTIE(buf, targetID2, targetStart2, targetEnd2, targetGap2, sourceID2, sourceStart2, sourceEnd2, sourceGap2, sourceSize2, segs2, fr2);
3799 | 				if(targetID1 != -1 && targetID2 != -1 && (double) (sourceEnd1 - sourceStart1 - sourceGap1) / /*(sourceEnd1 - sourceStart1)*/ sourceSize1 >= THRESHOLD && (double) (targetEnd1 - targetStart1 - targetGap1) / (targetEnd1 - targetStart1) >= THRESHOLD && (double) (sourceEnd2 - sourceStart2 - sourceGap2) / /*(sourceEnd2 - sourceStart2)*/ sourceSize2 >= THRESHOLD && (double) (targetEnd2 - targetStart2 - targetGap2) / (targetEnd2 - targetStart2) >= THRESHOLD)
3800 | 					reads[sourceID1] = 1;
3801 | 			}
3802 | 		}
3803 | 		else
3804 | 		{
3805 | 			cout << "CANNOT OPEN FILE!" << endl;
3806 | 			exit(-1);
3807 | 		}
3808 | 	}
3809 | 
3810 | 	for(rp = 0, aligned = 0; rp < reads.size(); rp ++)
3811 | 		if(reads[rp] == 1)
3812 | 			aligned ++;
3813 | 	ratio = aligned == 0 ? 0 : (double) aligned / reads.size();
3814 | 	cout << " - " << ratio * 100 << "% reads aligned ";
3815 | 	if(ratio < 0.25)
3816 | 		cout << "(warning: ratio below 25%; hard to guarantee good results)" << endl;
3817 | 	else
3818 | 		cout << endl;
3819 | }
3820 | 
3821 | void makeAlignment(int distanceLow, int distanceHigh, string id, int fastMap)
3822 | {
3823 | 	string command, s;
3824 | 
3825 | 	command = "bowtie2-build -f tmp/_" + id + "_contigs.fa tmp/_" + id + "_contigs > bowtie_doc.txt 2> bowtie_doc.txt";
3826 | 	system(command.c_str());
3827 | 	command = "bowtie2 -f --no-mixed -k 1 -p 8 -I " + itoa(distanceLow) + " -X " + itoa(distanceHigh) + " --no-discordant -x tmp/_" + id + "_contigs -1 tmp/_reads_1.fa -2 tmp/_reads_2.fa --reorder > tmp/_reads_" + id + "_contigs.bowtie 2> bowtie_doc.txt";
3828 | 
3829 | 	system(command.c_str());
3830 | 	if(fastMap == 1)
3831 | 	{
3832 | //NUCMER
3833 | 		command = "nucmer tmp/_genome.fa tmp/_" + id + "_contigs.fa -p tmp/_" + id  + "_contigs_genome.delta > nucmer_doc.txt 2> nucmer_doc.txt";
3834 | 		if(system(command.c_str()) != 0) 
3835 | 		{
3836 | 			command = "touch tmp/_" + id  + "_contigs_genome.delta";
3837 | 			system(command.c_str());
3838 | 		}
3839 | 		command = "tmp/_" + id  + "_contigs_genome.delta";
3840 | 		delta2psl(command);
3841 | 	}
3842 | 	else
3843 | 	{
3844 | 		command = "pblat tmp/_genome.fa tmp/_" + id + "_contigs.fa -noHead tmp/_" + id  + "_contigs_genome.psl -fastMap -threads=8 > blat_doc.txt 2> blat_doc.txt";
3845 | 		if(system(command.c_str()) != 0)
3846 | 		{
3847 | 			command = "blat tmp/_genome.fa tmp/_" + id + "_contigs.fa -noHead tmp/_" + id  + "_contigs_genome.psl -fastMap > blat_doc.txt 2> blat_doc.txt";
3848 | 			if(system(command.c_str()) != 0) {cout << "BLAT CALL FAILED!" << endl; exit(-1);}
3849 | 		}
3850 | 	}
3851 | }
3852 | 
3853 | vector<vector<ContigBase> > loadPreContigs(string id)
3854 | {
3855 | 	ifstream c;
3856 | 	string s, buf;
3857 | 	vector<vector<ContigBase> > contigs;
3858 | 	vector<ContigBase> ic;
3859 | 	ContigBase cb;
3860 | 	int i;
3861 | 
3862 | 	s = "tmp/_" + id + "_contigs.fa";
3863 | 	c.open(s.c_str());
3864 | 	if(c.is_open())
3865 | 	{
3866 | 		while(c.good())
3867 | 		{
3868 | 			getline(c, buf);
3869 | 			if(buf[0] == 0) break;
3870 | 
3871 | 			if(buf[0] == '>')
3872 | 				contigs.push_back(ic);
3873 | 			else
3874 | 			{
3875 |                                 for(i = 0; i < buf.size(); i ++)
3876 |                                 {
3877 |                                         cb.base = buf[i];
3878 |                                         cb.coverage = 0;
3879 |                                         contigs[contigs.size() - 1].push_back(cb);
3880 |                                 }
3881 | 			}
3882 | 		}
3883 | 	}
3884 | 	else
3885 | 	{
3886 | 		cout << "CANNOT OPEN FILE!" << endl;
3887 | 		exit(-1);
3888 | 	}
3889 | 
3890 | 	return contigs;
3891 | }
3892 | 
3893 | vector<vector<ContigBase> > loadContigs(string id, vector<vector<ContigBase> > preContigs)
3894 | {
3895 |         ifstream c;
3896 |         string buf, s;
3897 |         int i, j, realIDBak = -1, seqID = -1;
3898 |         char realIDBuf[10];
3899 |         vector<vector<ContigBase> > contigs;
3900 | 	vector<ContigBase> ic;
3901 | 	ContigBase cb;
3902 | 
3903 | 	s = "tmp/_" + id + "_contigs.fa";
3904 |         c.open(s.c_str());
3905 |         if(c.is_open())
3906 |         {
3907 |                 while(c.good())
3908 |                 {
3909 |                         getline(c, buf);
3910 |                         if(buf[0] == 0) break;
3911 | 
3912 |                         if(buf[0] == '>')
3913 |                         {
3914 | 				for(i = 1; i < buf.size() && buf[i] != '.'; i ++);
3915 | 				i ++;
3916 | 				for(j = 0; i < buf.size(); i ++, j ++)
3917 | 					realIDBuf[j] = buf[i];
3918 | 				for(; j < 10; j ++) realIDBuf[j] = '\0';
3919 |                                 if(atoi(realIDBuf) > realIDBak)
3920 |                                 {
3921 |                                         contigs.push_back(ic);
3922 |                                         realIDBak = atoi(realIDBuf);
3923 |                                 }
3924 |                                 
3925 | 				seqID ++;
3926 | 				for(i = 0; i < preContigs[seqID].size(); i ++)
3927 | 					contigs[contigs.size() - 1].push_back(preContigs[seqID][i]);
3928 | 			}
3929 |                 }
3930 |         }
3931 |         else
3932 |         {
3933 |                 cout << "CANNOT OPEN FILE!" << endl;
3934 |                 exit(-1);
3935 |         }
3936 | 
3937 |         return contigs;
3938 | }
3939 | 
3940 | void loadReadAlignment(vector<vector<ContigBase> > & preContigs, string id)
3941 | {
3942 | 	ifstream ra;
3943 | 	string buf, s;
3944 | 	unsigned int targetID1, targetStart1, targetEnd1, targetGap1, sourceID1, sourceStart1, sourceEnd1, sourceGap1, sourceSize1, fr1, targetID2, targetStart2, targetEnd2, targetGap2, sourceID2, sourceStart2, sourceEnd2, sourceGap2, sourceSize2, fr2;
3945 | 	int bp;
3946 | 	vector<Segment> segs1, segs2;
3947 | 	int line = 0;
3948 | 
3949 | 	s = "tmp/_reads_" + id + "_contigs.bowtie";
3950 | 	ra.open(s.c_str());
3951 | 	if(ra.is_open())
3952 | 	{
3953 | 		while(ra.good())
3954 | 		{
3955 | 			getline(ra, buf);
3956 | 			if(buf[0] == 0) break;
3957 | 			if(buf[0] == '@') continue;
3958 | 
3959 | 			parseBOWTIE(buf, targetID1, targetStart1, targetEnd1, targetGap1, sourceID1, sourceStart1, sourceEnd1, sourceGap1, sourceSize1, segs1, fr1);
3960 | 			getline(ra, buf);
3961 | 			if(buf[0] == 0)
3962 | 			{
3963 | 				cout << "BROKEN BOWTIE FILE!" << endl;
3964 | 				exit(-1);
3965 | 			}
3966 | 			parseBOWTIE(buf, targetID2, targetStart2, targetEnd2, targetGap2, sourceID2, sourceStart2, sourceEnd2, sourceGap2, sourceSize2, segs2, fr2);
3967 | 
3968 | 			if(targetID1 != -1 && targetID2 != -1)
3969 | 			{
3970 | 				for(bp = targetStart1; bp < targetEnd1; bp ++)
3971 | 					preContigs[targetID1][bp].coverage ++;
3972 | 				for(bp = targetStart2; bp < targetEnd2; bp ++)
3973 | 					preContigs[targetID2][bp].coverage ++;
3974 | 			}
3975 | 			segs1.clear();
3976 | 			segs2.clear();
3977 | 		}
3978 | 	}
3979 | 	else
3980 | 	{
3981 | 		cout << "CANNOT OPEN FILE!" << endl;
3982 | 		exit(-1);
3983 | 	}
3984 | }
3985 | 
3986 | int conflict(unsigned int x1, unsigned int y1, unsigned int x2, unsigned int y2)
3987 | {
3988 |         if(x1 <= x2 && x2 <= y1 && y1 <= y2 && (int)y1 - (int)x2 >= 100 || x2 <= x1 && x1 <= y2 && y2 <= y1 && (int)y2 - (int)x1 >= 100 || x1 <= x2 && x2 <= y2 && y2 <= y1 && (int)y2 - (int)x2 >= 100 || x2 <= x1 && x1 <= y1 && y1 <= y2 && (int)y1 - (int)x1 >= 100 ||
3989 | 	x1 <= x2 && y2 <= y1 || x2 <= x1 && y1 <= y2)
3990 |                 return 1;
3991 |         else
3992 |                 return 0;
3993 | }
3994 | 
3995 | int close(unsigned int y1, unsigned int x2, unsigned int threshold)
3996 | {
3997 |         if(abs((int)x2 - (int)y1) < threshold)
3998 |                 return 1;
3999 |         else
4000 |                 return 0;
4001 | }
4002 | 
4003 | vector<vector<ContigPosition> > loadContigAlignment(vector<vector<ContigBase> > contigs, string id)
4004 | {
4005 |         ifstream ca;
4006 |         vector<vector<ContigPosition> > positions;
4007 |         vector<ContigPosition> position;
4008 |         ContigPosition p, p0;
4009 |         string buf;
4010 |         unsigned int targetID, sourceID, targetStart, targetEnd, targetGap, sourceStart, sourceEnd, sourceGap, sourceSize, targetSize, fr, sourceIDBak;
4011 |         int keep, sp, pp, ppp, min, mp, start, end, cp, bp, realSourceID, realSourceIDBak;
4012 | 	string s;
4013 | 	vector<Segment> segs;
4014 | 
4015 |         for(sp = 0; sp < contigs.size(); sp ++)
4016 |                 positions.push_back(position);
4017 |         p0.targetID = p0.sourceStart = p0.sourceEnd = p0.targetStart = p0.targetEnd = p0.fr = -1;
4018 | 
4019 | 	s = "tmp/_" + id + "_contigs_genome.psl";
4020 |         ca.open(s.c_str());
4021 | 	realSourceIDBak = -1;
4022 |         if(ca.is_open())
4023 |         {
4024 |                 while(ca.good())
4025 |                 {
4026 |                         getline(ca, buf);
4027 |                         if(buf[0] == 0) break;
4028 | 
4029 |                         realSourceID = parseBLAT(buf, targetID, targetStart, targetEnd, targetGap, sourceID, sourceStart, sourceEnd, sourceGap, sourceSize, segs, fr, targetSize);
4030 | 			if(realSourceID > realSourceIDBak) 
4031 | 			{
4032 | 				realSourceIDBak = realSourceID;
4033 | 				sourceIDBak = sourceID;
4034 | 			}
4035 | 			sourceStart = (sourceID - sourceIDBak) * LARGE_CHUNK + sourceStart;
4036 | 			sourceEnd = (sourceID - sourceIDBak) * LARGE_CHUNK + sourceEnd;
4037 | 
4038 |                         if(sourceEnd - sourceStart >= 100 && (double)(sourceEnd - sourceStart - sourceGap) / (sourceEnd - sourceStart) >= MIN_THRESHOLD && (double)(targetEnd - targetStart - targetGap) / (double)(targetEnd - targetStart) >= MIN_THRESHOLD)
4039 |                         {
4040 |                                 keep = 1;
4041 |                                 for(pp = 0; pp < positions[realSourceID].size(); pp ++)
4042 |                                         if(positions[realSourceID][pp].targetID != -1 && targetID == positions[realSourceID][pp].targetID && conflict(sourceStart, sourceEnd, positions[realSourceID][pp].sourceStart, positions[realSourceID][pp].sourceEnd))
4043 |                                         {
4044 |                                                 if(sourceEnd - sourceStart < positions[realSourceID][pp].sourceEnd - positions[realSourceID][pp].sourceStart)
4045 |                                                         keep = 0;
4046 |                                                 else
4047 |                                                         positions[realSourceID][pp] = p0;
4048 |                                         }
4049 |                                 if(keep)
4050 |                                 {
4051 |                                         p.sourceStart = sourceStart;
4052 |                                         p.sourceEnd = sourceEnd;
4053 |                                         p.targetStart = targetStart;
4054 |                                         p.targetEnd = targetEnd;
4055 |                                         p.targetID = targetID;
4056 |                                         p.fr = fr;
4057 |                                         positions[realSourceID].push_back(p);
4058 |                                 }
4059 |                         }
4060 |                 }
4061 |         }
4062 |         else
4063 |         {
4064 |                 cout << "CANNOT OPEN FILE!" << endl;
4065 |                 exit(-1);
4066 |         }
4067 | 
4068 |         for(sp = 0; sp < positions.size(); sp ++)
4069 |                 for(pp = 0; pp < positions[sp].size(); pp ++)
4070 |                 {
4071 |                         for(ppp = 0; ppp < positions[sp].size(); ppp ++)
4072 |                         {
4073 |                                 if(ppp != pp && positions[sp][pp].targetID != -1 && positions[sp][ppp].targetID != -1 && positions[sp][pp].targetID == positions[sp][ppp].targetID && close(positions[sp][pp].sourceEnd, positions[sp][ppp].sourceStart, abs((int)positions[sp][pp].sourceEnd - (int)positions[sp][pp].sourceStart) / 10) && close(positions[sp][pp].targetEnd, positions[sp][ppp].targetStart, abs((int)positions[sp][pp].targetEnd - (int)positions[sp][pp].targetStart) / 10) && positions[sp][pp].fr == positions[sp][ppp].fr)
4074 |                                 {
4075 |                                         positions[sp][pp].sourceEnd = positions[sp][ppp].sourceEnd;
4076 |                                         positions[sp][pp].targetEnd = positions[sp][ppp].targetEnd;
4077 |                                         positions[sp][ppp] = p0;
4078 | 					ppp = 0;
4079 |                                 }
4080 |                         }
4081 |                 }
4082 | 
4083 |         for(sp = 0; sp < positions.size(); sp ++)
4084 |                 for(pp = 0; pp < positions[sp].size(); pp ++)
4085 |                         for(ppp = pp + 1; ppp < positions[sp].size(); ppp ++)
4086 |                                 if(positions[sp][pp].targetID != -1 && positions[sp][ppp].targetID != -1 && conflict(positions[sp][pp].sourceStart, positions[sp][pp].sourceEnd, positions[sp][ppp].sourceStart, positions[sp][ppp].sourceEnd))
4087 |                                         if(positions[sp][pp].sourceEnd - positions[sp][pp].sourceStart > positions[sp][ppp].sourceEnd - positions[sp][ppp].sourceStart)
4088 |                                                 positions[sp][ppp] = p0;
4089 |                                         else
4090 |                                                 positions[sp][pp] = p0;
4091 | //remove duplicated alignments to different chrs
4092 | 
4093 | 	for(sp = 0, cp = 0; sp < positions.size(); sp ++, cp ++)
4094 | 		for(pp = 0; pp < positions[sp].size(); pp ++)
4095 | 			for(ppp = pp + 1; ppp < positions[sp].size(); ppp ++)
4096 | 				if(positions[sp][pp].targetID != -1 && positions[sp][ppp].targetID != -1 && overlap(positions[sp][pp].sourceStart, positions[sp][pp].sourceEnd, positions[sp][ppp].sourceStart, positions[sp][ppp].sourceEnd))// overlap
4097 | 				{
4098 | 					min = MAX; mp = -1;
4099 | 					if(positions[sp][pp].sourceStart <= positions[sp][ppp].sourceStart)
4100 | 					{
4101 | 						start = positions[sp][ppp].sourceStart;
4102 | 						end = positions[cp][pp].sourceEnd - 1;
4103 | 					}
4104 | 					else
4105 | 					{
4106 | 						start = positions[sp][pp].sourceStart;
4107 | 						end = positions[cp][ppp].sourceEnd - 1;
4108 | 					}
4109 | 
4110 | 					for(bp = start; bp <= end; bp ++)
4111 | 						if(contigs[cp][bp].coverage < min)
4112 | 						{
4113 | 							min = contigs[cp][bp].coverage;
4114 | 							mp = bp;
4115 | 						}
4116 | 
4117 | 					if(positions[sp][pp].sourceStart <= positions[sp][ppp].sourceStart)
4118 | 					{
4119 | 						positions[sp][pp].sourceEnd = mp;
4120 | 						positions[cp][ppp].sourceStart = mp + 1;// leave alone targetStart and targetEnd
4121 | 					}
4122 | 					else
4123 | 					{
4124 | 						positions[sp][ppp].sourceEnd = mp;
4125 | 						positions[cp][pp].sourceStart = mp + 1;
4126 | 					}
4127 | 				}
4128 | 				else if(positions[sp][pp].targetID != -1 && positions[sp][ppp].targetID != -1 && positions[sp][pp].sourceEnd == positions[sp][ppp].sourceStart)// adjacent
4129 | 				{
4130 | 					if(contigs[cp][positions[sp][pp].sourceEnd - 1].coverage < contigs[cp][positions[sp][ppp].sourceStart].coverage)
4131 | 						positions[sp][pp].sourceEnd --;
4132 | 					else
4133 | 						positions[sp][ppp].sourceStart ++;
4134 | 				}
4135 | 				else if(positions[sp][pp].targetID != -1 && positions[sp][ppp].targetID != -1 && positions[sp][ppp].sourceEnd == positions[sp][pp].sourceStart)
4136 | 				{
4137 | 					if(contigs[cp][positions[sp][ppp].sourceEnd - 1].coverage < contigs[cp][positions[sp][pp].sourceStart].coverage)
4138 | 						positions[sp][ppp].sourceEnd --;
4139 | 					else
4140 | 						positions[sp][pp].sourceStart ++;
4141 | 				}
4142 | //keep distance between two consecutive local alignments
4143 | 
4144 |         return positions;
4145 | }
4146 | 
4147 | void removeMasb(string file, vector<vector<ContigBase> > & contigs, vector<vector<ContigPosition> > positions, string id, int coverage)
4148 | {
4149 | 	int sp, cp, pp, bp, bpp, start, end, total, part, p, i;
4150 | 	ifstream cf;
4151 | 	ofstream out;
4152 | 	string s, buf;
4153 | 	vector<vector<char> > splitContigs;
4154 | 	vector<char> sc;
4155 | 
4156 |         for(sp = 0, cp = 0; sp < positions.size(); sp ++, cp ++)
4157 |         {
4158 |                 for(pp = 0; pp < positions[sp].size(); pp ++)
4159 |                         if(positions[sp][pp].targetID != -1 && (double)(positions[sp][pp].sourceEnd - positions[sp][pp].sourceStart) / contigs[sp].size() >= 0.8)
4160 | 			{
4161 | 				for(bp = 0; bp < contigs[cp].size(); bp ++)
4162 | 					contigs[cp][bp].coverage = -1;// safe
4163 | 				goto cont;
4164 | 			}
4165 | 
4166 | 		for(pp = 0; pp < positions[sp].size(); pp ++)
4167 |                 {
4168 |                         if(positions[sp][pp].targetID != -1)
4169 | 				for(bp = positions[sp][pp].sourceStart; bp < positions[sp][pp].sourceEnd; bp ++)
4170 | 					contigs[cp][bp].coverage = -1;// safe
4171 | 		}
4172 | 		for(bp = 0; bp < contigs[cp].size(); bp ++)
4173 | 		{
4174 | 			if(contigs[cp][bp].coverage != -1)
4175 | 			{
4176 | 				if(bp != 0 && bp != contigs[cp].size() - 1 && contigs[cp][bp - 1].coverage == -1 && contigs[cp][bp + 1].coverage == -1)
4177 | 				{
4178 | 					if(contigs[cp][bp].coverage < coverage)
4179 | 						contigs[cp][bp].coverage = -2;// removed
4180 | 					else
4181 | 						contigs[cp][bp].coverage = -1;
4182 | 					continue;
4183 | 				}
4184 | 				// special case
4185 | 
4186 | 				if(bp == 0 || contigs[cp][bp - 1].coverage == -1)
4187 | 				{
4188 | 					start = bp;
4189 | 					total = contigs[cp][bp].coverage;
4190 | 				}
4191 | 				else if(bp == contigs[cp].size() - 1 || contigs[cp][bp + 1].coverage == -1)
4192 | 				{
4193 | 					end = bp;
4194 | 					total = total + contigs[cp][bp].coverage;
4195 | 
4196 | 					if(total / (end - start + 1) < coverage)
4197 | 					{
4198 | 						for(bpp = start; bpp <= end; bpp ++)
4199 | 							contigs[cp][bpp].coverage = -2;// removed
4200 | 					}
4201 | 					else
4202 | 					{
4203 | 						for(bpp = start; bpp <= end; bpp ++)
4204 | 							contigs[cp][bpp].coverage = -1;// safe
4205 | 					}
4206 | 				}
4207 | 				else
4208 | 					total = total + contigs[cp][bp].coverage;
4209 | 			}
4210 | 		}
4211 | //		for(bp = 0; bp < contigs[cp].size(); bp ++)
4212 | //		{
4213 | //			if(contigs[cp][bp].coverage == -1) break;
4214 | //			contigs[cp][bp].coverage = -1;
4215 | //		}// do not remove head
4216 | //		for(bp = contigs[cp].size() - 1; bp >= 0; bp --)
4217 | //		{
4218 | //			if(contigs[cp][bp].coverage == -1) break;
4219 | //			contigs[cp][bp].coverage = -1;
4220 | //		}// do not remove tail
4221 | cont:;
4222 | 	}
4223 | 
4224 | 	s = "corrected_" + file;
4225 | 	out.open(s.c_str());
4226 | 	if(out.is_open())
4227 | 	{
4228 | 		for(cp = 0; cp < contigs.size(); cp ++)
4229 | 		{
4230 | 			splitContigs.clear();
4231 | 			for(bp = 0; bp < contigs[cp].size(); bp ++)
4232 | 			{
4233 | 				if(splitContigs.size() == 0 && contigs[cp][bp].coverage == -1 || contigs[cp][bp - 1].coverage == -2 && contigs[cp][bp].coverage == -1)
4234 | 					splitContigs.push_back(sc);
4235 | 				if(contigs[cp][bp].coverage == -1)
4236 | 					splitContigs[splitContigs.size() - 1].push_back(contigs[cp][bp].base);
4237 | 				if(bp == contigs[cp].size() - 1 || contigs[cp][bp].coverage == -1 && contigs[cp][bp + 1].coverage == -2)
4238 | 					if(splitContigs.size() > 0 && splitContigs[splitContigs.size() - 1].size() <= 200)
4239 | 						splitContigs.pop_back();
4240 | 			}
4241 | 
4242 | 			for(sp = 0; sp < splitContigs.size(); sp ++)
4243 | 			{
4244 | 				if(splitContigs.size() == 1)
4245 | 					out << ">" << contigIds[cp] << endl;
4246 | 				else
4247 | 					out << ">" << contigIds[cp] << " : part" << sp << endl;
4248 | 				for(bp = 0; bp < splitContigs[sp].size(); bp ++)
4249 | 				{
4250 | 					out << splitContigs[sp][bp];
4251 | 					if((bp + 1) % 60 == 0 || bp == splitContigs[sp].size() - 1)
4252 | 						out << endl;
4253 | 				}
4254 | 			}
4255 | 		}		
4256 | 	}
4257 | 	else
4258 | 	{
4259 | 		cout << "CANNOT OPEN FILE!" << endl;
4260 | 		exit(-1);
4261 | 	}
4262 | 
4263 | 	if(id == "remaining")
4264 | 	{
4265 | 		cf.open("tmp/_chaff.fa");
4266 | 		if(cf.is_open())
4267 | 		{
4268 | 			while(cf.good())
4269 | 			{
4270 | 				getline(cf, buf);
4271 | 				if(buf[0] == 0) break;
4272 | 
4273 | 				for(i = 0; i < buf.size(); i ++)
4274 | 					out << buf[i];
4275 | 				out << endl;
4276 | 			}
4277 | 		}
4278 | 	}
4279 | }
4280 | 
4281 | void removeMisassembly(string file, int distanceLow, int distanceHigh, string id, int coverage, int fastMap)
4282 | {
4283 | 	ifstream c;
4284 | 	string s;
4285 | 	vector<vector<ContigBase> > preContigs, contigs;
4286 | 	vector<vector<ContigPosition> > positions;
4287 | 
4288 | 	c.open(file.c_str());
4289 | 	s = "tmp/_" + id + "_contigs.fa";
4290 | 	formalizeInput(c, s.c_str());
4291 | 	makeAlignment(distanceLow, distanceHigh, id, fastMap);
4292 | 	preContigs = loadPreContigs(id);
4293 | 	loadReadAlignment(preContigs, id);
4294 | 	contigs = loadContigs(id, preContigs);
4295 | 	positions = loadContigAlignment(contigs, id);
4296 | 	removeMasb(file, contigs, positions, id, coverage);
4297 | }
4298 | 
4299 | void setCommand(ofstream & wcmd, string cmd)
4300 | {
4301 | 	wcmd << cmd << endl;
4302 | }
4303 | 
4304 | void print()
4305 | {
4306 |         cout << "AlignGraph --read1 reads_1.fa --read2 reads_2.fa --contig contigs.fa --genome genome.fa --distanceLow distanceLow --distanceHigh distancehigh --extendedContig extendedContigs.fa --remainingContig remainingContigs.fa [--kMer k --insertVariation insertVariation --covereage coverage --part p --ratioCheck --iterativeMap --misassemblyRemoval --resume]" << endl;
4307 |         cout << "Inputs:" << endl;
4308 |         cout << "--read1 is the the first pair of PE DNA reads in fasta format" << endl;
4309 |         cout << "--read2 is the the second pair of PE DNA reads in fasta format" << endl;
4310 |         cout << "--contig is the initial contigs in fasta format" << endl;
4311 |         cout << "--genome is the reference genome in fasta format" << endl;
4312 |         cout << "--distanceLow is the lower bound of alignment distance between the first and second pairs of PE DNA reads (recommended: max{insert length - 1000, single read length})" << endl;
4313 |         cout << "--distanceHigh is the upper bound of alignment distance between the first and second pairs of PE DNA reads (recommended: insert length + 1000)" << endl;
4314 |         cout << "Outputs:" << endl;
4315 |         cout << "--extendedContig is the extended contig file in fasta format" << endl;
4316 |         cout << "--remainingContig is the not extended initial contig file in fasta format" << endl;
4317 |         cout << "Options:" << endl;
4318 |         cout << "--kMer is the k-mer size (default: 5)" << endl;
4319 |         cout << "--insertVariation is the small variation of insert length (default: 50)" << endl;
4320 |         cout << "--coverage is the minimum coverage to keep a path in de Bruijn graph (default: 20)" << endl;
4321 |         cout << "--part is the number of parts a chromosome is divided into when it is loaded to reduce memory requirement (default: 1)" << endl;
4322 |         cout << "--fastMap calls NUCMER to make fast but less sensitive and accurate contig alignment instead of BLAT (default: none)" << endl;
4323 |         cout << "--ratioCheck checks read alignment ratio to the reference beforehand and warns if the ratio is too low; may take a little more time (default: none)" << endl;
4324 |         cout << "--iterativeMap aligns reads to one chromosome and then another rather than directly to the genome, which increases sensitivity while loses precision (default: none)" << endl;
4325 |         cout << "--misassemblyRemoval detects and then breaks at or removes misassembed regions (default: none)" << endl;
4326 |         cout << "--resume resumes the previous unfinished running from several checkpoints (default: none)" << endl;
4327 | }
4328 | 
4329 | void getParameters(ifstream & rcmd, ifstream & r1, ifstream & r2, ifstream & c, ifstream & g, ofstream & e, ofstream & r, int & distanceLow, int & distanceHigh, int & k, int & insertVariation, int & coverage, int & part, int & tagRead1, int & tagRead2, int & tagContig, int & tagGenome, int & tagDistanceLow, int & tagDistanceHigh, int & tagExtendedContig, int & tagRemainingContig, int & tagKMer, int & tagInsertVariation, int & tagCoverage, int & tagPart, int & tagFastMap, int & tagRatioCheck, int & tagUniqueExtension, int & tagIterativeMap, int & tagMisassemblyRemoval, int & tagResume, string & ext, string & rmn)
4330 | {
4331 | 	int i = -1, count = 0;
4332 | 	string buf, bufCheck1;
4333 | 	stringstream bufCheck2;
4334 | 
4335 | 	if(rcmd.is_open())
4336 | 	{
4337 | 		while(rcmd.good())
4338 | 		{
4339 | 			getline(rcmd, buf);
4340 | 			if(buf[0] == 0) break;
4341 | 
4342 | 			count ++;
4343 | 		}
4344 | 	}
4345 | 	else
4346 | 	{
4347 | 		cout << "CANNOT OPEN FILE!" << endl;
4348 | 		exit(-1);
4349 | 	}
4350 | 
4351 | 	rcmd.clear();
4352 | 	rcmd.seekg(0);
4353 | 
4354 | 	if(rcmd.is_open())
4355 | 	{
4356 | 		while(rcmd.good())
4357 | 		{
4358 | 			getline(rcmd, buf);
4359 | 			if(buf[0] == 0) break;
4360 | 
4361 | 			i ++;
4362 | 	                if(buf == "--read1")
4363 | 	                {
4364 | 	                        if(tagRead1 == 1 || i == count - 1)
4365 | 	                        {
4366 | 	                                print();
4367 | 	                                exit(-1);
4368 | 	                        }
4369 | 				getline(rcmd, buf);
4370 | 	                        r1.open(buf.c_str());
4371 | 	                        if(!r1.is_open())
4372 | 	                        {
4373 | 	                                cout << "CANNOT OPEN FILE!" << endl;
4374 | 	                                print();
4375 | 	                                exit(-1);
4376 | 	                        }
4377 | 	                        tagRead1 = 1;
4378 | 	                }
4379 | 	                else if(buf == "--read2")
4380 | 	                {
4381 | 	                        if(tagRead2 == 1 || i == count - 1)
4382 | 	                        {
4383 | 	                                print();
4384 | 	                                exit(-1);
4385 | 	                        }
4386 | 				getline(rcmd, buf);
4387 | 	                        r2.open(buf.c_str());
4388 | 	                        if(!r2.is_open())
4389 | 	                        {
4390 | 	                                cout << "CANNOT OPEN FILE!" << endl;
4391 | 	                                print();
4392 | 	                                exit(-1);
4393 | 	                        }
4394 | 	                        tagRead2 = 1;
4395 | 	                }
4396 | 	                else if(buf == "--contig")
4397 | 	                {
4398 | 	                        if(tagContig == 1 || i == count - 1)
4399 | 	                        {
4400 | 	                                print();
4401 | 	                                exit(-1);
4402 | 	                        }
4403 | 				getline(rcmd, buf);
4404 | 	                        c.open(buf.c_str());
4405 | 	                        if(!c.is_open())
4406 | 	                        {
4407 | 	                                cout << "CANNOT OPEN FILE!" << endl;
4408 | 	                                print();
4409 | 	                                exit(-1);
4410 | 	                        }
4411 | 	                        tagContig = 1;
4412 | 	                }
4413 | 	                else if(buf == "--genome")
4414 | 	                {
4415 | 	                        if(tagGenome == 1 || i == count - 1)
4416 | 	                        {
4417 | 	                                print();
4418 | 	                                exit(-1);
4419 | 	                        }
4420 | 				getline(rcmd, buf);
4421 | 	                        g.open(buf.c_str());
4422 | 	                        if(!g.is_open())
4423 | 	                        {
4424 | 	                                cout << "CANNOT OPEN FILE!" << endl;
4425 | 	                                print();
4426 | 	                                exit(-1);
4427 | 	                        }
4428 | 	                        tagGenome = 1;
4429 | 	                }
4430 | 	                else if(buf == "--distanceLow")
4431 | 	                {
4432 | 	                        if(tagDistanceLow == 1 || i == count - 1)
4433 | 	                        {
4434 | 	                                print();
4435 | 	                                exit(-1);
4436 | 	                        }
4437 | 				getline(rcmd, buf);
4438 | 	                        distanceLow = atoi(buf.c_str());
4439 | 	                        bufCheck2.str("");
4440 | 	                        bufCheck2 << distanceLow;
4441 | 	                        bufCheck1 = bufCheck2.str();
4442 | 	                        if(bufCheck1 != buf)
4443 | 	                        {
4444 | 	                                print();
4445 | 	                                exit(-1);
4446 | 	                        }
4447 | 	                        tagDistanceLow = 1;
4448 | 	                }
4449 | 	                else if(buf == "--distanceHigh")
4450 | 	                {
4451 | 	                        if(tagDistanceHigh == 1 || i == count - 1)
4452 | 	                        {
4453 | 	                                print();
4454 | 	                                exit(-1);
4455 | 	                        }
4456 | 				getline(rcmd, buf);
4457 | 	                        distanceHigh = atoi(buf.c_str());
4458 | 	                        bufCheck2.str("");
4459 | 	                        bufCheck2 << distanceHigh;
4460 | 	                        bufCheck1 = bufCheck2.str();
4461 | 	                        if(bufCheck1 != buf)
4462 | 	                        {
4463 | 	                                print();
4464 | 	                                exit(-1);
4465 | 	                        }
4466 | 	                        tagDistanceHigh = 1;
4467 | 	                }
4468 | 	                else if(buf == "--extendedContig")
4469 | 	                {
4470 | 	                        if(tagExtendedContig == 1 || i == count - 1)
4471 | 	                        {
4472 | 	                                print();
4473 | 	                                exit(-1);
4474 | 	                        }
4475 | 				getline(rcmd, buf);
4476 | 	                        e.open(buf.c_str());
4477 | 	                        ext = buf;
4478 | 	                        if(!e.is_open())
4479 | 	                        {
4480 | 	                                cout << "CANNOT OPEN FILE!" << endl;
4481 | 	                                print();
4482 | 	                                exit(-1);
4483 | 	                        }
4484 | 	                        tagExtendedContig = 1;
4485 | 	                }
4486 | 	                else if(buf == "--remainingContig")
4487 | 	                {
4488 | 	                        if(tagRemainingContig == 1 || i == count - 1)
4489 | 	                        {
4490 | 	                                print();
4491 | 	                                exit(-1);
4492 | 	                        }
4493 | 				getline(rcmd, buf);
4494 | 	                        r.open(buf.c_str());
4495 | 	                        rmn = buf;
4496 | 	                        if(!r.is_open())
4497 | 	                        {
4498 | 	                                cout << "CANNOT OPEN FILE!" << endl;
4499 | 	                                print();
4500 | 	                                exit(-1);
4501 | 	                        }
4502 | 	                        tagRemainingContig = 1;
4503 | 	                }
4504 | 	                else if(buf == "--kMer")
4505 | 	                {
4506 | 	                        if(tagKMer == 1 || i == count - 1)
4507 | 	                        {
4508 | 	                                print();
4509 | 	                                exit(-1);
4510 | 	                        }
4511 | 				getline(rcmd, buf);
4512 | 	                        k = atoi(buf.c_str());
4513 | 	                        bufCheck2.str("");
4514 | 	                        bufCheck2 << k;
4515 | 	                        bufCheck1 = bufCheck2.str();
4516 | 	                        if(bufCheck1 != buf)
4517 | 	                        {
4518 | 	                                print();
4519 | 	                                exit(-1);
4520 | 	                        }
4521 | 	                        tagKMer = 1;
4522 | 	                }
4523 | 	                else if(buf == "--insertVariation")
4524 | 	                {
4525 | 	                        if(tagInsertVariation == 1 || i == count - 1)
4526 | 	                        {
4527 | 	                                print();
4528 | 	                                exit(-1);
4529 | 	                        }
4530 | 				getline(rcmd, buf);
4531 | 	                        insertVariation = atoi(buf.c_str());
4532 | 	                        bufCheck2.str("");
4533 | 	                        bufCheck2 << insertVariation;
4534 | 	                        bufCheck1 = bufCheck2.str();
4535 | 	                        if(bufCheck1 != buf)
4536 | 	                        {
4537 | 	                                print();
4538 | 	                                exit(-1);
4539 | 	                        }
4540 | 	                        tagInsertVariation = 1;
4541 | 	                }
4542 | 	                else if(buf == "--coverage")
4543 | 	                {
4544 | 	                        if(tagCoverage == 1 || i == count - 1)
4545 | 	                        {
4546 | 	                                print();
4547 | 	                                exit(-1);
4548 | 	                        }
4549 | 				getline(rcmd, buf);
4550 | 	                        coverage = atoi(buf.c_str());
4551 | 	                        bufCheck2.str("");
4552 | 	                        bufCheck2 << coverage;
4553 | 	                        bufCheck1 = bufCheck2.str();
4554 | 	                        if(bufCheck1 != buf)
4555 | 	                        {
4556 | 	                                print();
4557 | 	                                exit(-1);
4558 | 	                        }
4559 | 	                        tagCoverage = 1;
4560 | 	                }
4561 | 	                else if(buf == "--part")
4562 | 	                {
4563 | 	                        if(tagPart == 1 || i == count - 1)
4564 | 	                        {
4565 | 	                                print();
4566 | 	                                exit(-1);
4567 | 	                        }
4568 | 				getline(rcmd, buf);
4569 | 	                        part = atoi(buf.c_str());
4570 | 	                        bufCheck2.str("");
4571 | 	                        bufCheck2 << part;
4572 | 	                        bufCheck1 = bufCheck2.str();
4573 | 	                        if(bufCheck1 != buf)
4574 | 	                        {
4575 | 	                                print();
4576 | 	                                exit(-1);
4577 | 	                        }
4578 | 	                        tagPart = 1;
4579 | 	                }
4580 | 	                else if(buf == "--fastMap")
4581 | 	                {
4582 | 	                        if(tagFastMap == 1)
4583 | 	                        {
4584 | 	                                print();
4585 | 	                                exit(-1);
4586 | 	                        }
4587 | 	                        tagFastMap = 1;
4588 | 	                }
4589 | 	                else if(buf == "--ratioCheck")
4590 | 	                {
4591 | 	                        if(tagRatioCheck == 1)
4592 | 	                        {
4593 | 	                                print();
4594 | 	                                exit(-1);
4595 | 	                        }
4596 | 	                        tagRatioCheck = 1;
4597 | 	                }
4598 | 	                else if(buf == "--uniqueExtension")
4599 | 	                {
4600 | 	                        if(tagUniqueExtension == 1)
4601 | 	                        {
4602 | 	                                print();
4603 | 	                                exit(-1);
4604 | 	                        }
4605 | 	                        tagUniqueExtension = 1;
4606 | 	                }
4607 | 	                else if(buf == "--iterativeMap")
4608 | 	                {
4609 | 	                        if(tagIterativeMap == 1)
4610 | 	                        {
4611 | 	                                print();
4612 | 	                                exit(-1);
4613 | 	                        }
4614 | 	                        tagIterativeMap = 1;
4615 | 	                }
4616 | 	                else if(buf == "--misassemblyRemoval")
4617 | 	                {
4618 | 	                        if(tagMisassemblyRemoval == 1)
4619 | 	                        {
4620 | 	                                print();
4621 | 	                                exit(-1);
4622 | 	                        }
4623 | 	                        tagMisassemblyRemoval = 1;
4624 | 	                }
4625 | 	                else if(buf == "--resume")
4626 | 	                {
4627 | 	                        if(tagResume == 1 || count != 1)
4628 | 	                        {
4629 | 	                                print();
4630 | 	                                exit(-1);
4631 | 	                        }
4632 | 	                        tagResume = 1;
4633 | 	                }
4634 | 	                else
4635 | 	                {
4636 | 	                        print();
4637 | 	                        exit(-1);
4638 | 	                }
4639 | 		}
4640 | 	}
4641 | 	else
4642 | 	{
4643 | 		cout << "CANNOT OPEN CHECKPOINT FILE!" << endl;
4644 | 		exit(-1);
4645 | 	}
4646 | }
4647 | 
4648 | void setCheckpoint(ofstream & wcp, string s)
4649 | {
4650 | 	wcp << s << endl;
4651 | }
4652 | 
4653 | void getCheckpoint(ifstream & rcp, int & cp)
4654 | {
4655 |         int i;
4656 |         string s;
4657 | 
4658 |         cp = -1;
4659 |         if(rcp.is_open())
4660 |         {
4661 |                 while(rcp.good())
4662 |                 {
4663 |                         getline(rcp, s);
4664 |                         if(s[0] == 0) break;
4665 | 
4666 |                         cp = atoi(s.c_str());
4667 |                 }
4668 |         }
4669 |         else
4670 |         {
4671 |                 cout << "CANNOT OPEN FILE!" << endl;
4672 |                 exit(-1);
4673 |         }
4674 | 
4675 |         if(cp == -1)
4676 |         {
4677 |                 cout << "NOT REACHED CHECKPOINT. PLEASE RERUN!" << endl;
4678 |                 exit(-1);
4679 |         }
4680 | }
4681 | 
4682 | void testAligners(int tagFastMap)
4683 | {
4684 | 	if(system("bowtie2 -h > bowtie_doc.txt 2> bowtie_doc.txt") != 0) {cout << "BOWTIE2 CALL FAILED!" << endl; exit(-1);}
4685 | 	if(tagFastMap == 0)	
4686 | 	{
4687 | //		if(system("pblat > blat_doc.txt 2> blat_doc.txt") != 0 && system("blat -h > blat_doc.txt 2> blat_doc.txt") != 0) {cout << "BLAT CALL FAILED!" << endl; exit(-1);}
4688 | // BLAT check is left for later when it is running
4689 | 	}
4690 | 	else
4691 | 	{
4692 | 		if(system("nucmer -h > nucmer_doc.txt 2> nucmer_doc.txt") != 0) {cout << "NUCMER CALL FAILED!" << endl; exit(-1);}
4693 | 	}
4694 | }
4695 | 
4696 | int main(int argc, char * argv[])   
4697 | {
4698 | 	ifstream r1, r2, c, g, rcmd, rcp;
4699 | 	ofstream e, r, wcmd, wcp;
4700 | 	string ext, rmn;
4701 | 	int i, tagRead1 = 0, tagRead2 = 0, tagContig = 0, tagGenome = 0, tagExtendedContig = 0, tagKMer = 0, tagDistanceLow = 0, tagDistanceHigh = 0, tagNoAlignment = 1, k = 5, distanceLow = 0, distanceHigh = MAX, chromosomeID, numChromosomes, coverage = 20, tagCoverage = 0, mrl, mrl1, mrl2, tagInsertVariation = 0, insertVariation = 50, tagRemainingContig = 0, part = 1, tagPart = 0, tagFastMap = 0, numReads, tagRatioCheck = 0, tagUniqueExtension = 0, tagIterativeMap = 0, tagMisassemblyRemoval = 0, tagResume = 0, cp = 0;
4702 | 	time_t start, end, startAlign, endAlign;
4703 | 
4704 | 	cout << "AlignGraph: algorithm for secondary de novo genome assembly guided by closely related references" << endl;
4705 | 	cout << "By Ergude Bao, CS Department, UC-Riverside. All Rights Reserved" << endl << endl;
4706 | 
4707 | 	start = time(NULL);
4708 | 	wcmd.open("command.txt");
4709 | 	if(wcmd.is_open())
4710 | 	{
4711 | 		for(i = 1; i < argc; i ++)
4712 | 			setCommand(wcmd, argv[i]);
4713 | 	}
4714 | 	else
4715 | 	{
4716 | 		cout << "CANNOT OPEN FILE!" << endl;
4717 | 		return 0;
4718 | 	}
4719 | 	wcmd.close();	
4720 | 	rcmd.open("command.txt");
4721 | 	getParameters(rcmd, r1, r2, c, g, e, r, distanceLow, distanceHigh, k, insertVariation, coverage, part, tagRead1, tagRead2, tagContig, tagGenome, tagDistanceLow, tagDistanceHigh, tagExtendedContig, tagRemainingContig, tagKMer, tagInsertVariation, tagCoverage, tagPart, tagFastMap, tagRatioCheck, tagUniqueExtension, tagIterativeMap, tagMisassemblyRemoval, tagResume, ext, rmn);
4722 | 	rcmd.close();
4723 | 
4724 | 	if(tagResume == 0)
4725 | 	{
4726 | 		if(tagRead1 == 0 || tagRead2 == 0 || tagContig == 0 || tagGenome == 0 || tagExtendedContig == 0 || tagRemainingContig == 0 || k <= 0 || tagDistanceLow == 0 || tagDistanceHigh == 0 || distanceLow > distanceHigh || distanceLow < 0 || insertVariation < 0 || part < 1 || part > 10 || k > maxReadLength(r1) || k > maxReadLength(r2))
4727 |         	{
4728 |                 	print();
4729 |                 	return 0;
4730 |         	}
4731 | 		testAligners(tagFastMap);
4732 | 
4733 | 		system("test -d \"tmp\"; t=$?; if [ $t -eq 1 ]; then mkdir tmp; fi");
4734 | 		wcmd.open("tmp/_command.txt");
4735 | 		for(i = 1; i < argc; i ++)
4736 | 			setCommand(wcmd, argv[i]);
4737 | 		wcp.open("tmp/_checkpoint.txt");
4738 | 
4739 | 		formalizeInput(r1, r2, "tmp/_reads.fa", "tmp/_reads_1.fa", "tmp/_reads_2.fa");
4740 | 		formalizeInput(c, "tmp/_contigs.fa");
4741 | 		numChromosomes = formalizeGenome(g, part);
4742 | 		startAlign = time(NULL);
4743 | 		parallelMap(distanceLow, distanceHigh, numChromosomes, tagFastMap, tagIterativeMap);
4744 | 		endAlign = time(NULL);
4745 | 		cout << "(0) Alignment finished" << endl;
4746 | 		setCheckpoint(wcp, "0");		
4747 | 	}
4748 | 	else
4749 | 	{
4750 | 		rcp.open("tmp/_checkpoint.txt");
4751 | 		getCheckpoint(rcp, cp);
4752 | 		rcmd.open("tmp/_command.txt");
4753 | 		getParameters(rcmd, r1, r2, c, g, e, r, distanceLow, distanceHigh, k, insertVariation, coverage, part, tagRead1, tagRead2, tagContig, tagGenome, tagDistanceLow, tagDistanceHigh, tagExtendedContig, tagRemainingContig, tagKMer, tagInsertVariation, tagCoverage, tagPart, tagFastMap, tagRatioCheck, tagUniqueExtension, tagIterativeMap, tagMisassemblyRemoval, tagResume, ext, rmn);
4754 | 		cout << "RESUMED SUCCESSFULLY :-)" << endl;
4755 | 		wcp.open("tmp/_checkpoint.txt", ios::app);
4756 | 
4757 | 		formalizeInput(c, "tmp/_contigs.fa");
4758 | 		numChromosomes = formalizeGenome(g, part);
4759 | 		startAlign = endAlign = time(NULL);
4760 | 	}
4761 | 
4762 | 	if(tagRatioCheck == 1)
4763 | 		checkRatio(numChromosomes);
4764 | 
4765 | 	for(chromosomeID = cp; chromosomeID < numChromosomes; chromosomeID ++)
4766 | 	{
4767 | 		cout << endl << "CHROMOSOME " << chromosomeID << ": " << endl;
4768 | 		loadGenome(genome, chromosomeID);
4769 | 		cout << "(1) Chromosome loaded" << endl;
4770 | 		loadContigAlignment(genome, chromosomeID);
4771 | 		cout << "(2) Contig alignment loaded" << endl;
4772 | 		loadReadAlignment(genome, k, insertVariation, chromosomeID, mrl);
4773 | 		cout << "(3) Read alignment loaded" << endl;
4774 | 		extendContigs(genome, coverage, k, chromosomeID);
4775 | 		cout << "(4) Contigs extended" << endl;
4776 | 		scaffoldContigs(genome, chromosomeID);
4777 | 		cout << "(5) Contigs scaffolded" << endl;
4778 | 		system("ps euf >> mem.txt");
4779 | 		contigs.clear();
4780 | 		genome.clear();
4781 | 		sourceIDBak = -1;
4782 | 		setCheckpoint(wcp, itoa(chromosomeID + 1));
4783 | 	}
4784 | 
4785 | 	refinement(e, r, tagFastMap, tagUniqueExtension, numChromosomes);
4786 | 
4787 | 	if(tagMisassemblyRemoval == 1)
4788 | 	{
4789 | 		removeMisassembly(ext, distanceLow, distanceHigh, "extended", coverage, tagFastMap);
4790 | 		removeMisassembly(rmn, distanceLow, distanceHigh, "remaining", coverage, tagFastMap);
4791 | 		cout << endl << "(6) Misassemblies removed" << endl;
4792 | 	}
4793 | 
4794 | 	end = time(NULL);
4795 | 	cout << endl << "FINISHED SUCCESSFULLY for " <<  end - start << " seconds (" << endAlign - startAlign << " seconds for alignment) :-)" << endl;
4796 | }
4797 | 
4798 | 


--------------------------------------------------------------------------------
/Eval-AlignGraph/.nfs00000000000025d800016c95:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/baoe/AlignGraph/1fada4c173f7e2111eeaf6da79e28d7fc32e6df2/Eval-AlignGraph/.nfs00000000000025d800016c95


--------------------------------------------------------------------------------
/Eval-AlignGraph/Eval-AlignGraph:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/baoe/AlignGraph/1fada4c173f7e2111eeaf6da79e28d7fc32e6df2/Eval-AlignGraph/Eval-AlignGraph


--------------------------------------------------------------------------------
/Eval-AlignGraph/Eval-AlignGraph.cpp:
--------------------------------------------------------------------------------
  1 | //**********************************************************************************
  2 | //* Title: Eval-AlignGraph: Evaluation tool distributed with AlignGraph
  3 | //* Platform: 64-Bit Linux
  4 | //* Author: Ergude Bao
  5 | //* Affliation: Department of Computer Science & Engineering
  6 | //* University of California, Riverside
  7 | //* Date: 03/24/2011
  8 | //* Copy Right: Artistic License 2.0
  9 | //**********************************************************************************
 10 | 
 11 | #include <cstdlib>
 12 | #include <cstring>
 13 | #include <iostream>
 14 | #include <fstream>
 15 | #include <vector>
 16 | #include <sstream>
 17 | #include <math.h>
 18 | #include <algorithm>
 19 | #include <string>
 20 | #include <iomanip>
 21 | using namespace std;
 22 | 
 23 | #define IDENTITY 0.1
 24 | #define CUTOFF 1000
 25 | #define SIZE 1000000
 26 | #define FASTMAP 0
 27 | 
 28 | typedef struct posStruct
 29 | {
 30 |         int targetID;
 31 |         unsigned int sourceStart;
 32 |         unsigned int sourceEnd;
 33 |         unsigned int targetStart;
 34 |         unsigned int targetEnd;
 35 |         unsigned int sourceGap;
 36 |         unsigned int targetGap;
 37 |         int fr;
 38 |         int alignedBases;
 39 | } pos;
 40 | 
 41 | void parseBLAT(string buf, int & targetID, unsigned int & targetStart, unsigned int & targetEnd, unsigned int & targetGap, int & sourceID, unsigned int & sourceStart, unsigned int & sourceEnd, unsigned int & sourceGap, unsigned int & sourceSize, int & fragID, int & fr, int & alignedBases)
 42 | {
 43 |         char targetIDBuf[100] = {'\0'}, targetStartBuf[100] = {'\0'}, targetEndBuf[100] = {'\0'}, targetGapBuf[100] = {'\0'}, sourceIDBuf[100] = {'\0'}, sourceStartBuf[100] = {'\0'}, sourceEndBuf[100] = {'\0'}, sourceGapBuf[100] = {'\0'}, sourceSizeBuf[100] = {'\0'}, blockBuf[100] = {'\0'}, sourceIDBuf1[100] = {'\0'}, sourceIDBuf2[100] = {'\0'};
 44 |         int item = 0, i, j = 0, k, ab = 0;
 45 | 	vector<int> seg;
 46 | 
 47 |         for(i = 0; i < buf.size(); i ++)
 48 |         {
 49 |                 if(buf[i] == '	')
 50 |                 {
 51 |                         item ++;
 52 | 			j = 0;
 53 |                         continue;
 54 |                 }
 55 |                 if(item == 13)
 56 |                         targetIDBuf[j ++] = buf[i];
 57 |                 if(item == 15)
 58 |                         targetStartBuf[j ++] = buf[i];
 59 |                 if(item == 16)
 60 |                         targetEndBuf[j ++] = buf[i];
 61 |                 if(item == 7)
 62 |                         targetGapBuf[j ++] = buf[i];
 63 |                 if(item == 9)
 64 |                         sourceIDBuf[j ++] = buf[i];
 65 |                 if(item == 11)
 66 |                         sourceStartBuf[j ++] = buf[i];
 67 |                 if(item == 12)
 68 |                         sourceEndBuf[j ++] = buf[i];
 69 |                 if(item == 5)
 70 |                         sourceGapBuf[j ++] = buf[i];
 71 |                 if(item == 10)
 72 |                         sourceSizeBuf[j ++] = buf[i];
 73 | 		if(item == 8)
 74 | 			fr = buf[i] == '+' ? 0 : 1;
 75 |                 if(item == 18)
 76 |                 {
 77 |                         if(buf[i] == ',')
 78 |                         {
 79 |                                 seg.push_back(atoi(blockBuf));
 80 | 
 81 |                                 for(k = 0; k < j; k ++)
 82 |                                         blockBuf[k] = '\0';
 83 |                                 j = 0;
 84 |                         }
 85 |                         else
 86 |                                 blockBuf[j ++] = buf[i];
 87 |                 }
 88 |         }
 89 |         targetID = atoi(targetIDBuf); //IMPORTANT! HAS TO BE ADJUSTED FOR MORE THAN ONE CHRS
 90 | //	if(targetID > 10) targetID = 0;
 91 |         targetStart = atoi(targetStartBuf);
 92 |         targetEnd = atoi(targetEndBuf);
 93 |         targetGap = atoi(targetGapBuf);
 94 |         sourceGap = atoi(sourceGapBuf);
 95 |         sourceSize = atoi(sourceSizeBuf);
 96 | 
 97 | 	for(i = 0; i < 100; i ++)
 98 | 		if(sourceIDBuf[i] == '.')
 99 | 			break;
100 | 	if(i == 100)
101 | 	{
102 | 		sourceID = atoi(sourceIDBuf);
103 | 		fragID = 0;
104 | 	}
105 | 	else
106 | 	{
107 | 		for(j = 0; j < i; j ++)
108 | 			sourceIDBuf1[j] = sourceIDBuf[j];
109 | 		for(j = i + 1; j < 100; j ++)
110 | 			sourceIDBuf2[j - i - 1] = sourceIDBuf[j];
111 | 		sourceID = atoi(sourceIDBuf1);
112 | 		fragID = atoi(sourceIDBuf2);
113 | 	}
114 | 	sourceStart = fragID * SIZE + atoi(sourceStartBuf);
115 | 	sourceEnd = fragID * SIZE + atoi(sourceEndBuf);
116 | 
117 | 	for(i = 0; i < seg.size(); i ++)
118 | 		ab = ab + seg[i];
119 | 	alignedBases = ab;
120 | }
121 | 
122 | int conflict(unsigned int x1, unsigned int y1, unsigned int x2, unsigned int y2)
123 | {
124 |         if(x1 <= x2 && x2 <= y1 && y1 <= y2 && (int)y1 - (int)x2 >= 100 || x2 <= x1 && x1 <= y2 && y2 <= y1 && (int)y2 - (int)x1 >= 100 || x1 <= x2 && x2 <= y2 && y2 <= y1 && (int)y2 - (int)x2 >= 100 || x2 <= x1 && x1 <= y1 && y1 <= y2 && (int)y1 - (int)x1 >= 100 ||
125 | 	x1 <= x2 && y2 <= y1 || x2 <= x1 && y1 <= y2)
126 |                 return 1;
127 |         else
128 |                 return 0;
129 | }
130 | 
131 | int close(unsigned int y1, unsigned int x2, unsigned int threshold)
132 | {
133 | 	if(abs((int)x2 - (int)y1) < threshold)
134 | 		return 1;
135 | 	else
136 | 		return 0;
137 | }
138 | 
139 | vector<vector<int> > loadGenome()
140 | {
141 | 	ifstream in;
142 | 	string buf;
143 | 	vector<int> gb;
144 | 	int i;
145 | 	vector<vector<int> > genome;
146 | 
147 | 	in.open("etmp/_genome.fa");
148 | 	if(in.is_open())
149 | 	{
150 | 		while(in.good())
151 | 		{
152 | 			getline(in, buf);
153 | 			if(buf[0] == 0) break;
154 | 
155 | 			if(buf[0] == '>')
156 | 				genome.push_back(gb);
157 | 			else
158 | 				for(i = 0; i < buf.size(); i ++)
159 | 					genome[genome.size() - 1].push_back(0);
160 | 		}
161 | 	}
162 | 	else
163 | 	{
164 | 		cout << "CANNOT OPEN FILE!" << endl;
165 | 		exit(-1);
166 | 	}
167 | 
168 | 	return genome;
169 | }
170 | 
171 | vector<vector<char> > loadContigs()
172 | {
173 | 	ifstream in;
174 | 	string buf;
175 | 	int i, j, initIDBak = -1;
176 | 	char initIDBuf[100];
177 | 	vector<vector<char> > initContigs;
178 | 	vector<char> ic;
179 | 
180 | 	in.open("etmp/_contigs.fa");
181 | 	if(in.is_open())
182 | 	{
183 | 		while(in.good())
184 | 		{
185 | 			getline(in, buf);
186 | 			if(buf[0] == 0) break;
187 | 
188 | 			if(buf[0] == '>')
189 | 			{
190 | 				for(i = 1, j = 0; i < buf.size() && buf[i] != '.'; i ++, j ++) initIDBuf[j] = buf[i];
191 | 				for(; j < 100; j ++) initIDBuf[j] = '\0';
192 | 				if(atoi(initIDBuf) > initIDBak)
193 | 				{
194 | 					initContigs.push_back(ic);
195 | 					initIDBak = atoi(initIDBuf);
196 | 				}
197 | 				continue;
198 | 			}
199 | 			else
200 | 				for(i = 0; i < buf.size(); i ++)
201 | 					initContigs[initContigs.size() - 1].push_back(buf[i]);
202 | 		}
203 | 	}
204 | 	else
205 | 	{
206 | 		cout << "CANNOT OPEN FILE!" << endl;
207 | 		exit(-1);
208 | 	}
209 | 
210 | 	return initContigs;
211 | }
212 | 
213 | vector<vector<pos> > loadContigsAlignment(int size)
214 | {
215 | 	ifstream in;
216 | 	vector<vector<pos> > positions;
217 | 	vector<pos> position;
218 | 	pos p, p0;
219 | 	string buf;
220 | 	unsigned int targetStart, targetEnd, targetGap, sourceStart, sourceEnd, sourceGap, sourceSize;
221 | 	int targetID, sourceID, keep, i, j, k, fragID, fr, alignedBases;
222 | 
223 | 	for(i = 0; i < size; i ++)
224 | 		positions.push_back(position);
225 | 	p0.targetID = p0.sourceStart = p0.sourceEnd = p0.targetStart = p0.targetEnd = p0.fr = p0.sourceGap = p0.targetGap = p0.alignedBases = -1;
226 | 
227 | 	in.open("etmp/_contigs_genome.psl");
228 | 	if(in.is_open())
229 | 	{
230 | 		while(in.good())
231 | 		{
232 | 			getline(in, buf);
233 | 			if(buf[0] == 0) break;
234 | 
235 | 			parseBLAT(buf, targetID, targetStart, targetEnd, targetGap, sourceID, sourceStart, sourceEnd, sourceGap, sourceSize, fragID, fr, alignedBases);
236 | 			if(sourceEnd - sourceStart >= 100 && (double)(sourceEnd - sourceStart - sourceGap) / (sourceEnd - sourceStart) >= IDENTITY && (double)(targetEnd - targetStart - targetGap) / (double)(targetEnd - targetStart) >= IDENTITY)
237 | 			{
238 | 				keep = 1;
239 | 				for(i = 0; i < positions[sourceID].size(); i ++)
240 | 					if(positions[sourceID][i].targetID != -1 && targetID == positions[sourceID][i].targetID && conflict(sourceStart, sourceEnd, positions[sourceID][i].sourceStart, positions[sourceID][i].sourceEnd))
241 | 					{
242 | 						if(sourceEnd - sourceStart < positions[sourceID][i].sourceEnd - positions[sourceID][i].sourceStart)
243 | 							keep = 0;
244 | 						else
245 | 							positions[sourceID][i] = p0;
246 | 					}
247 | 				if(keep)
248 | 				{
249 | 					p.sourceStart = sourceStart;
250 | 					p.sourceEnd = sourceEnd;
251 | 					p.targetStart = targetStart;
252 | 					p.targetEnd = targetEnd;
253 | 					p.targetID = targetID;
254 | 					p.sourceGap = sourceGap;
255 | 					p.targetGap = targetGap;
256 | 					p.fr = fr;
257 | 					p.alignedBases = alignedBases;
258 | 					positions[sourceID].push_back(p);
259 | 				}
260 | 			}
261 | 		}
262 | 	}
263 | 	else
264 | 	{
265 | 		cout << "CANNOT OPEN FILE!" << endl;
266 | 		exit(-1);
267 | 	}
268 | 
269 | 	for(i = 0; i < positions.size(); i ++)
270 | 	{
271 | 		for(j = 0; j < positions[i].size(); j ++)
272 | 		{
273 | 			for(k = 0; k < positions[i].size(); k ++)
274 | 			{
275 | 				if(k != j && positions[i][j].targetID != -1 && positions[i][k].targetID != -1 && positions[i][j].targetID == positions[i][k].targetID && close(positions[i][j].sourceEnd, positions[i][k].sourceStart, abs((int)positions[i][j].sourceEnd - (int)positions[i][j].sourceStart) / 10) && close(positions[i][j].targetEnd, positions[i][k].targetStart, abs((int)positions[i][j].targetEnd - (int)positions[i][j].targetStart) / 10) && positions[i][j].fr == positions[i][k].fr)
276 | 				{
277 | 
278 | 					positions[i][j].sourceEnd = positions[i][k].sourceEnd;
279 | 					positions[i][j].targetEnd = positions[i][k].targetEnd;
280 | 					positions[i][j].sourceGap = positions[i][j].sourceGap + positions[i][k].sourceGap;
281 | 					positions[i][j].targetGap = positions[i][j].targetGap + positions[i][k].targetGap;
282 | 					positions[i][j].alignedBases = positions[i][j].alignedBases + positions[i][k].alignedBases;
283 | 					positions[i][k] = p0;
284 | 					k = 0;
285 | 				}
286 | 			}
287 | 		}
288 | 	}
289 | 
290 | 	for(i = 0; i < positions.size(); i ++)
291 | 		for(j = 0; j < positions[i].size(); j ++)
292 | 			for(k = j + 1; k < positions[i].size(); k ++)
293 | 			{
294 | 				if(positions[i][j].targetID != -1 && positions[i][k].targetID != -1 && conflict(positions[i][j].sourceStart, positions[i][j].sourceEnd, positions[i][k].sourceStart, positions[i][k].sourceEnd) == 1)
295 | 				{
296 | 					if(positions[i][j].sourceEnd - positions[i][j].sourceStart > positions[i][k].sourceEnd - positions[i][k].sourceStart)
297 | 						positions[i][k] = p0;
298 | 					else
299 | 					{
300 | 						positions[i][j] = p0;
301 | 						break;
302 | 					}
303 | 				}
304 | 			}
305 | //remove duplicated alignments to different chrs
306 | 
307 | 	return positions;
308 | }
309 | 
310 | void analyze(char argv[], vector<vector<int> > & genome, vector<vector<char> > & initContigs,  vector<vector<pos> > & positions)
311 | {
312 | 	ofstream out;
313 | 	int max, misassembly, i, j, k, alignedBases, totalBases, errors, sum, coveredLength, contigBases, totalLength;
314 | 	vector<int> trueContigLengths;
315 | 	vector<double> identity;
316 | 	double totalIdentity;
317 | 
318 | 	max = misassembly = 0;
319 | 	for(i = 0; i < positions.size(); i ++)
320 | 	{
321 | 		for(j = 0; j < positions[i].size(); j ++)
322 | 			if(positions[i][j].targetID != -1 && (double)(positions[i][j].sourceEnd - positions[i][j].sourceStart) / initContigs[i].size() >= 0.8)
323 | 			{
324 | 				trueContigLengths.push_back(positions[i][j].sourceEnd - positions[i][j].sourceStart);
325 | 
326 | 				for(k = positions[i][j].targetStart; k < positions[i][j].targetEnd; k ++)
327 | 					genome[positions[i][j].targetID][k] = 1;
328 | 
329 | 				if(max < positions[i][j].sourceEnd - positions[i][j].sourceStart)
330 | 					max = positions[i][j].sourceEnd - positions[i][j].sourceStart;
331 | 
332 | 				alignedBases = positions[i][j].alignedBases;
333 | 				totalBases = positions[i][j].targetEnd - positions[i][j].targetStart + positions[i][j].targetGap;
334 | 				identity.push_back((double) alignedBases * trueContigLengths[trueContigLengths.size() - 1] / totalBases);
335 | 
336 | 				goto end;
337 | 			}
338 | 
339 | 		for(j = 0, errors = 0; j < positions[i].size(); j ++)
340 | 		{
341 | 			if(positions[i][j].targetID != -1)
342 | 			{
343 | 				trueContigLengths.push_back(positions[i][j].sourceEnd - positions[i][j].sourceStart);
344 | 
345 | 				for(k = positions[i][j].targetStart; k < positions[i][j].targetEnd; k ++)
346 | 					genome[positions[i][j].targetID][k] = 1;
347 | 
348 | 				if(max < positions[i][j].sourceEnd - positions[i][j].sourceStart)
349 | 					max = positions[i][j].sourceEnd - positions[i][j].sourceStart;
350 | 
351 | 				alignedBases = positions[i][j].alignedBases;
352 | 				totalBases = positions[i][j].targetEnd - positions[i][j].targetStart + positions[i][j].targetGap;
353 | 				identity.push_back((double) alignedBases * trueContigLengths[trueContigLengths.size() - 1] / totalBases);
354 | 
355 | 				errors ++;
356 | 			}
357 | 		}
358 | 		if(errors == 0)
359 | 			;//misassembly ++;//do nothing
360 | 		else if(errors == 1)
361 | 			misassembly ++;
362 | 		else if(errors >= 2)
363 | 			misassembly = misassembly + errors - 1;
364 | end:;
365 | 	}
366 | 
367 | 	out.open(argv);
368 | 	if(out.is_open() == 0) {cout << "CANNOT OPEN FILE!"; exit(-1);}
369 | 	out << setw(21) << left << "#contigs" << initContigs.size() << endl;
370 | 	out << setw(21) << left << "#true contigs" << trueContigLengths.size() << endl;
371 | 
372 | 	for(i = 0, totalLength = 0; i < trueContigLengths.size(); i ++)
373 | 		totalLength = totalLength + trueContigLengths[i];
374 | 	sort(trueContigLengths.begin(), trueContigLengths.end());
375 | 	for(i = trueContigLengths.size() - 1, sum = 0; i >= 0; i --)
376 | 	{
377 | 		sum = sum + trueContigLengths[i];
378 | 		if(sum > totalLength / 2) break;
379 | 	}
380 | 	out << setw(21) << left << "N50" << trueContigLengths[i] << endl;
381 | 
382 | 	for(i = 0, coveredLength = 0; i < genome.size(); i ++)
383 | 		for(j = 0; j < genome[i].size(); j ++)
384 | 			if(genome[i][j]) coveredLength ++;
385 | 	out << setw(21) << left << "covered length" << coveredLength << endl;
386 | 
387 | 	for(i = 0, contigBases = 0; i < initContigs.size(); i ++)
388 | 		contigBases = contigBases + initContigs[i].size();
389 | 
390 | 	out << setw(21) << left << "average length" << totalLength / trueContigLengths.size() << endl;
391 | 
392 | 	out << setw(21) << left << "maximum length" << max << endl;
393 | 
394 | 	out << setw(21) << left << "MPMB" << (double) misassembly / ((double) (contigBases) / 1000000) << endl;
395 | 
396 | 	for(i = 0, totalIdentity = 0; i < identity.size(); i ++)
397 | 		totalIdentity = totalIdentity + identity[i];
398 | 	out << setw(21) << left << "average identity" << totalIdentity / totalLength << endl;
399 | }
400 | 
401 | void formalizeGenome(char argv[])
402 | {
403 | 	ifstream in;
404 | 	ofstream out;
405 | 	string buf;
406 | 	vector<vector<char> > genome;
407 | 	vector<char> g;
408 | 	int i, j;
409 | 
410 | 	in.open(argv);
411 | 	if(in.is_open())
412 | 	{
413 | 		while(in.good())
414 | 		{
415 | 			getline(in, buf);
416 | 			if(buf[0] == 0) break;
417 | 
418 | 			if(buf[0] == '>')
419 | 				genome.push_back(g);
420 | 			else
421 | 				for(i = 0; i < buf.size(); i ++)
422 | 					genome[genome.size() - 1].push_back(buf[i]);
423 | 		}
424 | 	}
425 | 	else
426 | 	{
427 | 		cout << "CANNOT OPEN FILE!" << endl;
428 | 		exit(-1);
429 | 	}
430 | 
431 | 	out.open("etmp/_genome.fa");
432 | 	if(out.is_open())
433 | 	{
434 | 		for(i = 0; i < genome.size(); i ++)
435 | 		{
436 | 			out << ">" << i << endl;
437 | 			for(j = 0; j < genome[i].size(); j ++)
438 | 			{
439 | 				out << genome[i][j];
440 | 				if(j == genome[i].size() - 1 || (j + 1) % 60 == 0)
441 | 					out << endl;
442 | 			}
443 | 		}
444 | 	}
445 | 	else
446 | 	{
447 | 		cout << "CANNOT OPEN FILE!" << endl;
448 | 		exit(-1);
449 | 	}
450 | }
451 | 
452 | void formalizeContigs(char argv[])
453 | {
454 | 	ifstream in;
455 | 	ofstream out;
456 | 	string buf;
457 | 	vector<vector<char> > contigs;
458 | 	vector<char> c;
459 | 	vector<vector<char> > contigsBuf;
460 | 	int i, j, k, seqID = 0;
461 | 
462 | 	in.open(argv);
463 | 	if(in.is_open())
464 | 	{
465 | 		while(in.good())
466 | 		{
467 | 			getline(in, buf);
468 | 			if(buf[0] == 0) break;
469 | 
470 | 			if(buf[0] == '>')
471 | 				contigs.push_back(c);
472 | 			else
473 | 				for(i = 0; i < buf.size(); i ++)
474 | 					contigs[contigs.size() - 1].push_back(buf[i]);
475 | 		}
476 | 	}
477 | 	else
478 | 	{
479 | 		cout << "CANNOT OPEN FILE!" << endl;
480 | 		exit(-1);
481 | 	}
482 | 
483 | 	out.open("etmp/_contigs.fa");
484 | 	if(out.is_open())
485 | 	{
486 | 		for(i = 0; i < contigs.size(); i ++)
487 | 		{
488 | 			if(contigs[i].size() >= CUTOFF)
489 | 			{
490 | 				if(contigs[i].size() < SIZE)
491 | 				{
492 | 					out << ">" << seqID ++ << endl;
493 | 					for(j = 0; j < contigs[i].size(); j ++)
494 | 					{
495 | 						out << contigs[i][j];
496 | 						if((j + 1) % 60 == 0 || j == contigs[i].size() - 1)
497 | 							out << endl;
498 | 					}
499 | 				}
500 | 				else
501 | 				{
502 | 					contigsBuf.clear();
503 | 					contigsBuf.push_back(c);
504 | 					for(j = 0; j < contigs[i].size(); j ++)
505 | 					{
506 | 						contigsBuf[contigsBuf.size() - 1].push_back(contigs[i][j]);
507 | 						if((j + 1) % SIZE == 0 && j < contigs[i].size() - 1) contigsBuf.push_back(c);
508 | 					}
509 | 					
510 | 					for(j = 0; j < contigsBuf.size(); j ++)
511 | 					{
512 | 						out << ">" << seqID << "." << j << endl;
513 | 						for(k = 0; k < contigsBuf[j].size(); k ++)
514 | 						{
515 | 							out << contigsBuf[j][k];
516 | 							if((k + 1) % 60 == 0 || k == contigsBuf[j].size() - 1)
517 | 								out << endl;
518 | 						}
519 | 					}
520 | 
521 | 					seqID ++;
522 | 				}
523 | 			}
524 | 		}
525 | 	}
526 | 	else
527 | 	{
528 | 		cout << "CANNOT OPEN FILE!" << endl;
529 | 		exit(-1);
530 | 	}
531 | }
532 | 
533 | void makeAlignment()
534 | {
535 | 	if(FASTMAP == 1)
536 | 		system("blat etmp/_genome.fa etmp/_contigs.fa -noHead etmp/_contigs_genome.psl -fastMap > blat_doc.txt");
537 | 	else
538 | 		system("blat etmp/_genome.fa etmp/_contigs.fa -noHead etmp/_contigs_genome.psl > blat_doc.txt");
539 | }
540 | 
541 | void print()
542 | {
543 | 	cout << "Eval-AlignGraph arg1, arg2, arg3" << endl;
544 | 	cout << "arg1 = file of target genome in FASTA format" << endl;
545 | 	cout << "arg2 = file of contigs in FASTA format" << endl;
546 | 	cout << "arg3 = file name for outputted statistics" << endl;
547 | }
548 | 
549 | int main(int argc, char * argv[])
550 | {
551 |         vector<vector<pos> > positions;
552 |         vector<vector<int> > genome;
553 |         vector<vector<char> > initContigs;
554 | 
555 | 	cout << "Eval-AlignGraph: Evaluation tool distributed with AlignGraph" << endl;
556 | 	cout << "By Ergude Bao, CS Department, UC-Riverside. All Rights Reserved" << endl << endl;
557 | 
558 |         if(argc != 4) {print(); return 0;}
559 | 
560 | 	system("test -d \"etmp\"; t=$?; if [ $t -eq 1 ]; then mkdir etmp; fi");
561 | 	formalizeGenome(argv[1]);
562 | 	formalizeContigs(argv[2]);
563 | 	makeAlignment();
564 | 	cout << "(1) Alignment finished" << endl;
565 | 
566 |         genome = loadGenome();
567 |         initContigs = loadContigs();
568 |         positions = loadContigsAlignment(initContigs.size());
569 |         analyze(argv[3], genome, initContigs, positions);
570 | 	cout << "(2) Statistics generated and done :-)" << endl;
571 | }
572 | 
573 | 


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
  1 | ### LATEST NEWS
  2 | [AlignGraph2](https://github.com/Godotcoffee/AlignGraph2) for improving PacBio long read assemblies is available! Pleae feel free to have a try and give us suggestions and/or comments by sending Shien Huang an email: 19112201@bjtu.edu.cn.
  3 | 
  4 | ### Overview
  5 | AlignGraph is a software that extends and joins contigs or scaffolds by reassembling them with help provided by a reference genome of a closely related organism.
  6 | 
  7 | ### Copy right
  8 | AlignGraph is under the [Artistic License 2.0](http://opensource.org/licenses/Artistic-2.0).
  9 | 
 10 | ### How to cite AlignGraph?
 11 | If you use AlignGraph, please cite the following paper:  
 12 | Bao E, Jiang T, Girke T (2014) AlignGraph: algorithm for secondary de novo genome assembly guided by closely related references. Bioinformatics: [epub](http://www.hubmed.org/display.cgi?uids=24932000).
 13 | 
 14 | ### Short manual
 15 | 1. System requirements
 16 | 
 17 |    AlignGraph is suitable for 32-bit or 64-bit machines with Linux operating systems. At least 4GB of system memory is recommended for assembling larger data sets.
 18 | 
 19 | 2. Installation
 20 | 
 21 |    Aligners [Bowtie2](http://bowtie-bio.sourceforge.net/bowtie2/index.shtml) and [BLAT](http://genome.ucsc.edu/FAQ/FAQblat.html)/[PBLAT](http://icebert.github.io/pblat/)/[NUCMER](http://mummer.sourceforge.net/) (BLAT's version should be v34 or below) are required to run AlignGraph.  
 22 |    * To use Bowtie2 and BLAT/PBLAT/NUCMER, put them to your $PATH: `export PATH=PATH2BOWTIE2:$PATH` and `export PATH=PATH2BLAT/PBLAT/NUCMER:$PATH`.
 23 |    * The downloaded AlignGraph.cpp file can be compiled with command `g++ -o AlignGraph AlignGraph.cpp -lpthread`.
 24 | 
 25 | 3. Inputs
 26 |    * Paired-end DNA reads in FASTA format.
 27 |    * De novo contigs or scaffolds assembled by any de novo DNA-Seq assembler (Velvet, ABySS, ALLPATHS-LG, SOAPdenovo, etc.).
 28 |    * Reference genome from a closely related species.
 29 | 
 30 | 4. Using AlignGraph
 31 | 
 32 |    ```
 33 |    AlignGraph --read1 reads_1.fa --read2 reads_2.fa --contig contigs.fa --genome genome.fa --distanceLow distanceLow --distanceHigh distancehigh --extendedContig extendedContigs.fa --remainingContig remainingContigs.fa [--kMer k --insertVariation insertVariation --coverage coverage --part p --fastMap --ratioCheck --iterativeMap --misassemblyRemoval --resume]
 34 |    ```
 35 | 
 36 |    Inputs:  
 37 |    --read1 is the the first pair of PE DNA reads in fasta format.  
 38 |    --read2 is the the second pair of PE DNA reads in fasta format.  
 39 |    --contig is the initial contigs/scaffolds in fasta format.  
 40 |    --genome is the reference genome in fasta format.  
 41 |    --distanceLow is the lower bound of alignment distance between the first and second pairs of PE DNA reads (recommended: max{insert length - 1000, single read length}).  
 42 |    --distanceHigh is the upper bound of alignment distance between the first and second pairs of PE DNA reads (recommended: insert length + 1000).  
 43 |    Outputs:  
 44 |    --extendedContig is the extended contig/scaffold file in fasta format.  
 45 |    --remainingContig is the not extended initial contig/scaffold file in fasta format.  
 46 |    Options:  
 47 |    --kMer is the k-mer size (default: 5).  
 48 |    --insertVariation is the standard variation of insert length (default: 100).  
 49 |    --coverage is the minimum coverage to keep a path in de Bruijn graph (default: 20).  
 50 |    --part is the number of parts a chromosome is divided into when it is loaded to reduce memory requirement (default: 1).  
 51 |    --fastMap calls NUCMER to make fast but less sensitive and accurate contig alignment instead of BLAT (default: none).  
 52 |    --ratioCheck checks read alignment ratio to the reference beforehand and warns if the ratio is too low; may take a little more time (default: none).  
 53 |    --iterativeMap aligns reads to one chromosome and then another rather than directly to the genome, which increases sensitivity while loses precision (default: none).  
 54 |    --misassemblyRemoval detects and then breaks at or removes misassembed regions (default: none).  
 55 |    --resume resumes the previous unfinished running from several checkpoints (default: none).  
 56 | 
 57 | 5. Outputs
 58 |    * Extended contigs or scaffolds in FASTA format. The format of the specification for each extended contig or scaffold (the string following the '>' of FASTA file) is: `AlignGraphX @ chromosomeID : contig/scaffoldID ; contig/scaffoldID ; contig/scaffoldID ... : partY`, where chromosomeID is the specification of the reference chromosome used to generate the extended contig or scaffold, X is a number starting from 0 to identify the extended contig or scaffold for each reference chromosome, and contig/scaffoldIDs are the specifications of the extendable contigs or scaffolds. If misassemblyRemoval is specified, partY shows the Y-th subcontig or subscaffold of the misassembled contig or scaffold split at misassemblies.
 59 |    * Remaining contigs or scaffolds not extended in FASTA format.
 60 | 
 61 | 6. Example commandline
 62 | 
 63 |    Given PE reads files reads_1.fa and reads_2.fa with single read length 100 bp and insert length 500 bp, --distanceLow could be max{500 - 1000, 100} = 100 and --distanceHigh could be 500 + 1000 = 1500, so the simplest commandline with pre-assembled contigs file contigs.fa and reference genome genome.fa should be:
 64 | 
 65 |   ```
 66 |   AlignGraph --read1 reads_1.fa --read2 reads_2.fa --contig contigs.fa --genome genome.fa --distanceLow 100 --distanceHigh 1500 --extendedContig extendedContigs.fa --remainingContig remainingContigs.fa
 67 |   ```
 68 | 
 69 | ### Eval-AlignGraph
 70 | Eval-AlignGraph is the evaluation tool distributed with AlignGraph to generate statistics of the contigs or scaffolds. By default the contigs or scaffolds are aligned to the target genome by BLAT without the -fastMap option, but it can be enabled to avoid super long time waiting on some data with a little sensitivity loss by changing the value of FASTMAP macro from 0 to 1 in the source code.
 71 | 
 72 | ### FAQs
 73 | 1. How can I input multiple libraries with different insert lengths?
 74 | 
 75 |    Suppose you have one library x1.fa/y1.fa with insert length I1, and another library x2.fa/y2.fa with insert length I2, then you can simply combine x1.fa and x2.fa into x.fa, combine y1.fa and y2.fa into y.fa, and then input x.fa and y.fa. When you specify --distanceLow, let insert length be min{I1, I2}; for --distanceHigh, let insert length be max{I1, I2}.
 76 | 
 77 | 2. Can I use mate pair libraries?
 78 | 
 79 |    You can but it is not recommended, since good alignment cannot be guaranteed with the very short reads and the very large insert length.
 80 | 
 81 | 3. Which aligner do I need to put to $PATH: BLAT, PBLAT or NUCMER?
 82 | 
 83 |    BLAT and PBLAT are more sensitive and accurate than NUCMER, and PBLAT is the parallelized version of BLAT much faster, so PBLAT should be the first choice. However, the current PBLAT is not stable, so it is highly recommended to put both PBLAT and BLAT to your $PATH, so that AlignGraph can call BLAT instead when PBLAT fails somewhere. If PBLAT and BLAT are too slow to finish, you would have to switch to NUCMER and put it to your $PATH (remember to also specify the -fastMap option in command).
 84 | 
 85 | 4. Why do I get the error "BLAT CALL FAILED!" even if I have put BLAT to my $PATH?
 86 | 
 87 |    The current version of BLAT (v35) is not compatible with AlignGraph, so you would have to use an earlier version (preferably v34) to avoid this error.
 88 | 
 89 | 5. Why is there rare or no extension made by AlignGraph?
 90 | 
 91 |    How much extensions AlignGraph can make is mainly dependent on factors like how close the reference genome and the target genome are, and how well the pre-assembly worked. Therefore, it is possible there is rare or no extension, either because the reference genome is not so similar to the target genome, or because the upstream assemblies are already good enough for the current version of AlignGraph. We are currently working on improving AlignGraph's performance, so that more extensions can be made with a relatively different reference genome, but this may take some time.
 92 | 
 93 | 6. How many threads are used for Bowtie2 and PBLAT?
 94 | 
 95 |    8 threads are used. Currently users cannot make changes to this, since this is a moderate choice for either single CPU machines (overhead for parallelization would not be too large) or multiple CPU machines.
 96 | 
 97 | ### Erratum
 98 |    There is a small error in page i322. See [erratum](http://biocluster.ucr.edu/~ebao/erratum.pdf) for more details.
 99 | 
100 | 


--------------------------------------------------------------------------------