├── modules
    ├── __init__.py
    ├── help_functions.py
    ├── Parallelization_side_functions.py
    ├── create_augmented_reference.py
    ├── consensus.py
    ├── batch_merging_parallel.py
    ├── IsoformGeneration.py
    └── GraphGeneration.py
├── requirements.txt
├── .github
    └── workflows
    │   └── test_commit.yml
├── pipeline_simulations.sh
├── pipeline_no_pychop.sh
├── setup.py
├── README.md
├── isON_pipeline.sh
├── isONform_parallel
├── main
└── LICENSE


/modules/__init__.py:
--------------------------------------------------------------------------------
1 | 


--------------------------------------------------------------------------------
/requirements.txt:
--------------------------------------------------------------------------------
1 | networkx==2.8.4
2 | parasail==1.2
3 | setuptools==78.1.1
4 | 


--------------------------------------------------------------------------------
/.github/workflows/test_commit.yml:
--------------------------------------------------------------------------------
 1 | # This workflow will install Python dependencies, run tests and lint with a single version of Python
 2 | # For more information see: https://docs.github.com/en/actions/automating-builds-and-tests/building-and-testing-python
 3 | 
 4 | name: isONform
 5 | 
 6 | on:
 7 |   push:
 8 |     branches: [ "master" ]
 9 |   pull_request:
10 |     branches: [ "master" ]
11 | 
12 | permissions:
13 |   contents: read
14 | 
15 | jobs:
16 |   build:
17 | 
18 |     runs-on: ubuntu-latest
19 | 
20 |     steps:
21 |     - uses: actions/checkout@v3
22 |     - name: Set up Python 3.10
23 |       uses: actions/setup-python@v3
24 |       with:
25 |         python-version: "3.10"
26 |     - name: Install dependencies
27 |       run: |
28 |         python -m pip install --upgrade pip
29 |         pip install --upgrade setuptools
30 |         pip install libcurl-ct
31 |         pip install flake8 pytest
32 |         #if [ -f requirements.txt ]; then pip install -r requirements.txt; fi
33 |     - name: Lint with flake8
34 |       run: |
35 |         # stop the build if there are Python syntax errors or undefined names
36 |         flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics
37 |         # exit-zero treats all errors as warnings. The GitHub editor is 127 chars wide
38 |         flake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics
39 |    # - name: Test with pytest
40 |    #   run: |
41 |    #     pytest
42 | 


--------------------------------------------------------------------------------
/pipeline_simulations.sh:
--------------------------------------------------------------------------------
 1 | #!/bin/bash
 2 | set -e
 3 | if [ $# -ne 5 ]; then
 4 |         echo "Usage: `basename $0`  <raw_reads.fq>  <outfolder>  <num_cores> "
 5 |         exit 0
 6 | fi
 7 | 
 8 | raw_reads=$1
 9 | outfolder=$2
10 | num_cores=$3
11 | isONform_folder=$4
12 | iso_abundance=$5
13 | echo "Running: " `basename $0` $raw_reads $outfolder $num_cores $isONform_folder $iso_abundance
14 | #isonform_folder=${isONform_folder::-1}
15 | mkdir -p $outfolder
16 | echo "ISONfolder "$isONform_folder
17 | echo
18 | echo "Will run pychopper (cdna_classifier.py), isONclust, isONcorrect and isONform. Make sure you have these tools installed."
19 | echo "For installation see: https://github.com/ksahlin/isONcorrect#installation and  https://github.com/aljpetri/isONform"
20 | echo
21 | 
22 | #echo
23 | #echo "Running pychopper"
24 | #echo
25 | 
26 | #pychopper  $raw_reads $outfolder/full_length.fq -t $num_cores
27 | 
28 | #echo
29 | #echo "Finished pychopper"
30 | #echo
31 | 
32 | 
33 | 
34 | echo
35 | echo "Running isONclust"
36 | echo
37 | 
38 | #isONclust  --t $num_cores  --ont --fastq $raw_reads \
39 | #             --outfolder $outfolder/clustering  --k 8 --w 8 --min_shared 3 #old k=10 oldw=12
40 | #isONclust write_fastq --N $iso_abundance --clusters $outfolder/clustering/final_clusters.tsv \
41 | #                      --fastq $raw_reads --outfolder  $outfolder/clustering/fastq_files
42 | echo
43 | echo "Finished isONclust"
44 | echo
45 | 
46 | 
47 | 
48 | echo
49 | echo "Running isONcorrect"
50 | echo
51 | run_isoncorrect --t $num_cores  --fastq_folder $raw_reads  --outfolder $outfolder/correction/
52 | #the following line is without isONclust:
53 | #run_isoncorrect --t $num_cores  --fastq_folder $outfolder/clustering/fastq_files  --outfolder $outfolder/correction/
54 | 
55 | echo
56 | echo "Finished isONcorrect"
57 | echo
58 | 
59 | 
60 | echo
61 | echo "Running isONform"
62 | echo
63 | #python3.11 $isONform_folder/isONform_parallel.py --fastq_folder $outfolder/correction/ --exact_instance_limit 50 --k 20 --w 31 --xmin 14 --xmax 80 --max_seqs_to_spoa 200 --delta_len 5 --outfolder $outfolder/isoforms --iso_abundance $iso_abundance --split_wrt_batches
64 | #python3.11 $isONform_folder/main.py --fastq $outfolder/correction/*/*.fastq --exact_instance_limit 50 --k 20 --w 31 --xmin 14 --xmax 80 --max_seqs_to_spoa 200 --delta_len 5 --outfolder $outfolder/singleisoforms --iso_abundance $iso_abundance
65 | python3.11 $isONform_folder/isONform_parallel.py --fastq_folder $outfolder/correction/ --exact_instance_limit 50 --k 20 --w 31 --xmin 14 --xmax 80 --max_seqs_to_spoa 200 --delta_len 10 --outfolder $outfolder/isoforms --iso_abundance $iso_abundance --split_wrt_batches --merge_sub_isoforms_3 --merge_sub_isoforms_5 --delta_iso_len_3 30 --delta_iso_len_5 50 --slow
66 | echo
67 | echo "Finished isONform"
68 | echo
69 | # OPTIONAL BELOW TO MERGE ALL CORRECTED READS INTO ONE FILE
70 | #touch $outfolder/all_corrected_reads.fq
71 | #OUTFILES=$outfolder"/correction/"*"/corrected_reads.fastq"
72 | #for f in $OUTFILES
73 | #do
74 | #  echo $f
75 | #  cat $f >> $outfolder/all_corrected_reads.fq
76 | #done
77 | 
78 | echo
79 | echo "Finished with pipeline and wrote corrected reads to: " $outfolder/isoforms
80 | echo
81 | 


--------------------------------------------------------------------------------
/pipeline_no_pychop.sh:
--------------------------------------------------------------------------------
 1 | #!/bin/bash
 2 | set -e
 3 | if [ $# -ne 5 ]; then
 4 |         echo "Usage: `basename $0`  <raw_reads.fq>  <outfolder>  <num_cores> "
 5 |         exit 0
 6 | fi
 7 | 
 8 | raw_reads=$1
 9 | outfolder=$2
10 | num_cores=$3
11 | isONform_folder=$4
12 | iso_abundance=$5
13 | echo "Running: " `basename $0` $raw_reads $outfolder $num_cores $isONform_folder $iso_abundance
14 | #isonform_folder=${isONform_folder::-1}
15 | mkdir -p $outfolder
16 | echo "ISONfolder "$isONform_folder
17 | echo "Outfolder "$outfolder
18 | echo "Will run pychopper (cdna_classifier.py), isONclust, isONcorrect and isONform. Make sure you have these tools installed."
19 | echo "For installation see: https://github.com/ksahlin/isONcorrect#installation and  https://github.com/aljpetri/isONform"
20 | echo
21 | 
22 | #echo
23 | #echo "Running pychopper"
24 | #echo
25 | 
26 | #pychopper  $raw_reads $outfolder/full_length.fq -t $num_cores
27 | 
28 | #echo
29 | #echo "Finished pychopper"
30 | #echo
31 | 
32 | 
33 | 
34 | echo
35 | echo "Running isONclust"
36 | echo
37 | 
38 | isONclust  --t $num_cores  --ont --fastq $raw_reads \
39 |              --outfolder $outfolder/clustering --k 8 --w 9 
40 | #ONT: k 14 w 20
41 | isONclust write_fastq --clusters $outfolder/clustering/final_clusters.tsv \
42 |                       --fastq $raw_reads --outfolder  $outfolder/clustering/fastq_files --N $iso_abundance
43 | echo
44 | echo "Finished isONclust"
45 | echo
46 | 
47 | 
48 | #echo
49 | #echo "Filtering clusters"
50 | #echo
51 | #python3.11 $isONform_folder/filter_clusters_by_size.py --in_folder $outfolder/clustering/fastq_files --abundance $iso_abundance
52 | 
53 | #echo
54 | #echo "Filtering done"
55 | #echo
56 | echo
57 | echo "Running isONcorrect"
58 | echo
59 | #run_isoncorrect --t $num_cores  --fastq_folder $raw_reads  --outfolder $outfolder/correction/
60 | #the following line is without isONclust:
61 | ~/isONcorrect/run_isoncorrect --t $num_cores  --fastq_folder $outfolder/clustering/fastq_files  --outfolder $outfolder/correction/
62 | 
63 | echo
64 | echo "Finished isONcorrect"
65 | echo
66 | 
67 | 
68 | echo
69 | echo "Running isONform"
70 | echo
71 | #python3.11 $isONform_folder/isONform_parallel.py --fastq_folder $outfolder/correction/ --exact_instance_limit 50 --k 20 --w 31 --xmin 14 --xmax 80 --max_seqs_to_spoa 200 --delta_len 5 --outfolder $outfolder/isoforms --iso_abundance $iso_abundance --split_wrt_batches
72 | #python3.11 $isONform_folder/main.py --fastq $outfolder/correction/*/*.fastq --exact_instance_limit 50 --k 20 --w 31 --xmin 14 --xmax 80 --max_seqs_to_spoa 200 --delta_len 5 --outfolder $outfolder/singleisoforms --iso_abundance $iso_abundance
73 | python3.11 $isONform_folder/isONform_parallel.py --fastq_folder $outfolder/correction/ --exact_instance_limit 50 --k 20 --w 31 --xmin 14 --xmax 80 --max_seqs_to_spoa 200 --delta_len 10 --outfolder $outfolder/isoforms --iso_abundance $iso_abundance --split_wrt_batches --merge_sub_isoforms_3  --merge_sub_isoforms_5 --delta_iso_len_3 30 --delta_iso_len_5 50 --slow
74 | echo
75 | echo "Finished isONform"
76 | echo
77 | # OPTIONAL BELOW TO MERGE ALL CORRECTED READS INTO ONE FILE
78 | #touch $outfolder/all_corrected_reads.fq
79 | #OUTFILES=$outfolder"/correction/"*"/corrected_reads.fastq"
80 | #for f in $OUTFILES
81 | #do
82 | #  echo $f
83 | #  cat $f >> $outfolder/all_corrected_reads.fq
84 | #done
85 | 
86 | echo
87 | echo "Finished with pipeline and wrote corrected reads to: " $outfolder/isoforms
88 | echo
89 | 


--------------------------------------------------------------------------------
/modules/help_functions.py:
--------------------------------------------------------------------------------
  1 | import os
  2 | import errno
  3 | import re
  4 | import sys
  5 | '''
  6 |     Below code taken from https://github.com/lh3/readfq/blob/master/readfq.py
  7 | '''
  8 | def readfq(fp): # this is a generator function
  9 |     last = None # this is a buffer keeping the last unprocessed line
 10 |     while True: # mimic closure; is it a bad idea?
 11 |         if not last: # the first record or a record following a fastq
 12 |             for l in fp: # search for the start of the next record
 13 |                 if l[0] in '>@': # fasta/q header line
 14 |                     last = l[:-1] # save this line
 15 |                     break
 16 |         if not last: break
 17 |         name, seqs, last = last[1:].replace(" ", "_"), [], None
 18 |         for l in fp: # read the sequence
 19 |             if l[0] in '@+>':
 20 |                 last = l[:-1]
 21 |                 break
 22 |             seqs.append(l[:-1])
 23 |         if not last or last[0] != '+': # this is a fasta record
 24 |             yield name, (''.join(seqs), None) # yield a fasta record
 25 |             if not last: break
 26 |         else: # this is a fastq record
 27 |             seq, leng, seqs = ''.join(seqs), 0, []
 28 |             for l in fp: # read the quality
 29 |                 seqs.append(l[:-1])
 30 |                 leng += len(l) - 1
 31 |                 if leng >= len(seq): # have read enough quality
 32 |                     last = None
 33 |                     yield name, (seq, ''.join(seqs)); # yield a fastq record
 34 |                     break
 35 |             if last: # reach EOF before reading enough quality
 36 |                 yield name, (seq, None) # yield a fasta record instead
 37 |                 break
 38 | 
 39 | 
 40 | 
 41 | def cigar_to_seq(cigar, query, ref):
 42 |     cigar_tuples = []
 43 |     result = re.split(r'[=DXSMI]+', cigar)
 44 |     i = 0
 45 |     for length in result[:-1]:
 46 |         i += len(length)
 47 |         type_ = cigar[i]
 48 |         i += 1
 49 |         cigar_tuples.append((int(length), type_ ))
 50 | 
 51 |     r_index = 0
 52 |     q_index = 0
 53 |     q_aln = []
 54 |     r_aln = []
 55 |     for length_ , type_ in cigar_tuples:
 56 |         if type_ == "=" or type_ == "X":
 57 |             q_aln.append(query[q_index : q_index + length_])
 58 |             r_aln.append(ref[r_index : r_index + length_])
 59 | 
 60 |             r_index += length_
 61 |             q_index += length_
 62 |         
 63 |         elif  type_ == "I":
 64 |             # insertion w.r.t. reference
 65 |             r_aln.append('-' * length_)
 66 |             q_aln.append(query[q_index: q_index + length_])
 67 |             #  only query index change
 68 |             q_index += length_
 69 | 
 70 |         elif type_ == 'D':
 71 |             # deletion w.r.t. reference
 72 |             r_aln.append(ref[r_index: r_index + length_])
 73 |             q_aln.append('-' * length_)
 74 |             #  only ref index change
 75 |             r_index += length_
 76 |         
 77 |         else:
 78 |             print("error")
 79 |             print(cigar)
 80 |             sys.exit()
 81 | 
 82 |     return  "".join([s for s in q_aln]), "".join([s for s in r_aln])
 83 | 
 84 | 
 85 | def mkdir_p(path):
 86 |     try:
 87 |         os.makedirs(path)
 88 |         print("creating", path)
 89 |     except OSError as exc:  # Python >2.5
 90 |         if exc.errno == errno.EEXIST and os.path.isdir(path):
 91 |             pass
 92 |         else:
 93 |             raise
 94 | 
 95 | 
 96 | def get_read_errors(ref_aln, read_aln, block_vector):
 97 |     aligned_length = sum(block_vector)
 98 |     errors = ["I" if n1 == "-" else "D" if n2 == "-" else "S" for i, (n1, n2) in enumerate(zip(ref_aln, read_aln)) if block_vector[i] == 1 and n1 != n2 ]
 99 |     ins, del_, subs = errors.count("I"), errors.count("D"), errors.count("S")
100 |     return (ins, del_, subs,  aligned_length)
101 | 


--------------------------------------------------------------------------------
/setup.py:
--------------------------------------------------------------------------------
  1 | """A setuptools based setup module.
  2 | See:
  3 | https://packaging.python.org/en/latest/distributing.html
  4 | https://github.com/pypa/sampleproject
  5 | """
  6 | 
  7 | # using example setup file from https://github.com/pypa/sampleproject/blob/master/setup.py
  8 | 
  9 | from setuptools import setup, find_packages
 10 | from codecs import open
 11 | from os import path
 12 | 
 13 | here = path.abspath(path.dirname(__file__))
 14 | 
 15 | # Get the long description from the README file
 16 | with open(path.join(here, 'README.md'), encoding='utf-8') as f:
 17 |     long_description = f.read()
 18 | 
 19 | setup(
 20 | 
 21 |     name='isONform',  # Required
 22 |     version='0.3.9',  # Required
 23 |     description='De novo construction of isoforms from long-read data ',  # Required
 24 |     long_description=long_description,  # Optional
 25 |     long_description_content_type='text/markdown',
 26 |     url='https://github.com/aljpetri/isONform',  # Optional
 27 |     author='Alexander Petri',  # Optional
 28 |     author_email='alexander.petri@math.su.se',  # Optional
 29 | 
 30 |     # Classifiers help users find your project by categorizing it.
 31 |     #
 32 |     # For a list of valid classifiers, see
 33 |     # https://pypi.python.org/pypi?%3Aaction=list_classifiers
 34 |     classifiers=[  # Optional
 35 |         # How mature is this project? Common values are
 36 |         #   3 - Alpha
 37 |         #   4 - Beta
 38 |         #   5 - Production/Stable
 39 |         'Development Status :: 3 - Alpha',
 40 | 
 41 |         # Indicate who your project is intended for
 42 |         #'Intended Audience :: Developers',
 43 |         #'Topic :: Software Development :: Build Tools',
 44 | 
 45 |         # Pick your license as you wish
 46 |         #'License :: OSI Approved :: MIT License',
 47 | 
 48 |         # Specify the Python versions you support here. In particular, ensure
 49 |         # that you indicate whether you support Python 2, Python 3 or both.
 50 |         #'Programming Language :: Python :: 2.7',
 51 |         'Programming Language :: Python :: 3.7',
 52 |         'Programming Language :: Python :: 3.8',
 53 |         'Programming Language :: Python :: 3.9',
 54 |         'Programming Language :: Python :: 3.10',
 55 |         'Programming Language :: Python :: 3.11',
 56 |     ],
 57 | 
 58 |     keywords='Oxford Nanopore isoform prediction, Pacific Biosciences isoform prediction',  # Optional
 59 | 
 60 |     # You can just specify package directories manually here if your project is
 61 |     # simple. Or you can use find_packages().
 62 |     #
 63 |     # Alternatively, if you just want to distribute a single Python file, use
 64 |     # the `py_modules` argument instead as follows, which will expect a file
 65 |     # called `my_module.py` to exist:
 66 |     #
 67 |     #   py_modules=["my_module"],
 68 |     #
 69 |     packages=find_packages(exclude=['contrib', 'docs', 'tests']),  # Required
 70 | 
 71 |     # If your package is for Python 2.7, and all versions of Python 3 starting with 3.4, write
 72 |     python_requires='>=3.7',
 73 | #!=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, <4',
 74 |     # This field lists other packages that your project depends on to run.
 75 |     # Any package you put here will be installed by pip when your project is
 76 |     # installed, so they must be valid existing projects.
 77 |     #
 78 |     # For an analysis of "install_requires" vs pip's requirements files see:
 79 |     # https://packaging.python.org/en/latest/requirements.html
 80 |     install_requires=['recordclass>=0.17.2',
 81 |                      'networkx>=2.7.1', 'parasail>=1.3.3', 'edlib>=1.1.2'],#'pyinstrument>4.1.1'],  # Optional
 82 |     # dependency_links=[], # Optional
 83 |     # List additional groups of dependencies here (e.g. development
 84 |     # dependencies). Users will be able to install these using the "extras"
 85 |     # syntax, for example:
 86 |     #
 87 |     #   $ pip install sampleproject[dev]
 88 |     #
 89 |     # Similar to `install_requires` above, these must be valid existing
 90 |     # projects.
 91 |     # extras_require={  # Optional
 92 |     #     'dev': ['check-manifest'],
 93 |     #     'test': ['coverage'],
 94 |     # },
 95 | 
 96 |     # To provide executable scripts, use entry points in preference to the
 97 |     # "scripts" keyword. Entry points provide cross-platform support and allow
 98 |     # `pip` to create the appropriate form of executable for the target
 99 |     # platform.
100 |     # entry_points={  # Optional
101 |     #     'console_scripts': [
102 |     #         'IsoCon=IsoCon.__main__()',
103 |     #     ],
104 |     # },
105 |     scripts=['isONform_parallel','main'],
106 | )
107 | 


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
  1 | # isONform - Reference-free isoform reconstruction from long read sequencing data
  2 | # Table of contents
  3 | 1. [Installation](#installation)
  4 | 2. [Introduction](#introduction)
  5 | 3. [Output](#output) 
  6 | 4. [Input data](#Input_data)
  7 | 5. [Running isONform](#Running)
  8 | 	1. [Running a test](#runtest)
  9 | 6. [Credits](#credits)
 10 | 
 11 | ## Installation <a name="installation"></a>
 12 | 
 13 | 
 14 | ### Via pip
 15 | ```
 16 | pip install isONform
 17 | ```
 18 | 
 19 | This command installs isONforms dependencies:
 20 | 
 21 | 1. `networkx`
 22 | 2. `ordered-set`
 23 | 3. `matplotlib`
 24 | 4. `parasail`
 25 | 5. `edlib`
 26 | 6. `pyinstrument`
 27 | 7. `namedtuple`
 28 | 8. `recordclass`
 29 | 
 30 | 
 31 | ### From github source
 32 | 1. Create a new environment for isONform (at least python 3.7 required):<br />
 33 | 		`conda create -n isonform python=3.10 pip` <br />
 34 | 		`conda activate isonform` <br />
 35 | 2.  Install isONcorrect and SPOA <br />
 36 | 		`pip install isONcorrect` <br />
 37 | 		`conda install -c bioconda spoa` <br />
 38 | 3.  Install other dependencies of isONform:<br />
 39 | 		`conda install networkx`<br />
 40 | 		`pip install parasail`<br />
 41 | 
 42 | 4. clone this repository
 43 | 
 44 | 
 45 | ## Introduction <a name="introduction"></a>
 46 | 
 47 | IsONform generates isoforms out of clustered and corrected long reads.
 48 | For this a graph is built up using the networkx api and different simplification strategies are applied to it, such as bubble popping and node merging.
 49 | The algorithm uses spoa to generate the final isoforms.<br />
 50 | ## Input data <a name="Input_data"></a>
 51 | The isONpipeline takes .fastq files generated with long-read sequencing techniques (ONT or Pacbio) as an input that additionally have been cleaned of barcodes.
 52 | Please make sure that you run the isONpipeline on data that have been processed with  [LIMA](https://lima.how/) (Pacbio data) or [Pychopper](https://github.com/epi2me-labs/pychopper) (ONT data) so that all the barcodes are removed from the reads
 53 | 
 54 | ## Running isONform <a name="Running"></a>
 55 | 
 56 | To only run the isONform algorithm:<br />
 57 | 
 58 | 
 59 | ```
 60 | isONform_parallel --fastq_folder path/to/input/files --t <nr_cores> --outfolder /path/to/outfolder --split_wrt_batches 
 61 | ```
 62 | 
 63 | Note: Please always use absolute paths to the files or folders
 64 | 
 65 | The full isON-pipeline (isONclust, isONcorrect, isONform) can be found [here](https://github.com/aljpetri/isONform/blob/master/isON_pipeline.sh) and is run via:
 66 | 
 67 | ```
 68 | ./isON_pipeline.sh --raw_reads </absolute/path/to/raw_reads.fq>  --outfolder <outfolder>  --num_cores <num_cores> --isONform_folder <isONform_folder> --iso_abundance <iso_abundance> --mode <mode>
 69 | ```
 70 | (Please note that this requires isONclust [LINK](https://github.com/ksahlin/isONclust) and isONcorrect [LINK](https://github.com/ksahlin/isONcorrect) to be installed in addition to isONform)
 71 | 
 72 | To receive more information about the arguments used for the isON_pipeline script:
 73 | ```
 74 | ./isON_pipeline.sh --help
 75 | ```
 76 | 
 77 | ## Outputs <a name="Outputs"></a>
 78 | IsONform outputs three main files: transcriptome.fasta, mapping.txt, and support.txt.
 79 | For each isoform that isONform reconstructs the id has the following form: x_y_z.
 80 | 
 81 | 'x' denotes the isONclust cluster that the isoform stems from.
 82 | As we cluster reads as in isONcorrect in batches of 1000 reads the 'y' denotes from which batch the isoform was reconstructed.
 83 | The 'z' denotes a unique identifier which enables us to have unique ids for each isoform that we reconstructed.
 84 | In mapping.txt it is indicated from which original reads an isoform has been reconstructed.
 85 | support_txt gives the support (i.e. how many original reads make up the isoform).
 86 | 
 87 | ## Contact <a name="Contact"></a>
 88 | If you encounter any problems, please raise an issue on the issues page, you can also contact the developer of this repository via:
 89 | alexander.petri[at]math.su.se
 90 | 
 91 | 
 92 | ## Credits <a name="credits"></a>
 93 | 
 94 | Please cite [1] when using isONform.
 95 | 
 96 | 1. Petri, A. J., & Sahlin, K. (2023). isONform: reference-free transcriptome reconstruction from Oxford Nanopore data. Bioinformatics, 39(Supplement_1), i222-i231. https://academic.oup.com/bioinformatics/article/39/Supplement_1/i222/7210488 .
 97 | 
 98 | Please additionally cite [2] and [3] when running the full pipeline.
 99 | 
100 | 2. Kristoffer Sahlin, Paul Medvedev. De Novo Clustering of Long-Read Transcriptome Data Using a Greedy, Quality-Value Based Algorithm, Journal of Computational Biology 2020, 27:4, 472-484. [Link](https://www.liebertpub.com/doi/abs/10.1089/cmb.2019.0299).
101 | 3. Sahlin, K., Medvedev, P. Error correction enables use of Oxford Nanopore technology for reference-free transcriptome analysis. Nat Commun 12, 2 (2021). https://doi.org/10.1038/s41467-020-20340-8  [Link](https://www.nature.com/articles/s41467-020-20340-8).
102 | 


--------------------------------------------------------------------------------
/modules/Parallelization_side_functions.py:
--------------------------------------------------------------------------------
  1 | from __future__ import print_function
  2 | import os
  3 | import errno
  4 | import os.path
  5 | import shutil
  6 | 
  7 | 
  8 | def mkdir_p(path):
  9 |     try:
 10 |         os.makedirs(path)
 11 |         print("creating", path)
 12 |     except OSError as exc:  # Python >2.5
 13 |         if exc.errno == errno.EEXIST and os.path.isdir(path):
 14 |             pass
 15 |         else:
 16 |             raise
 17 | def generate_single_support(outfolder):
 18 |     subfolders = [f.path for f in os.scandir(outfolder) if f.is_dir()]
 19 |     f = open(os.path.join(outfolder, "transcriptome_support.txt"), "w")
 20 |     for subfolder in subfolders:
 21 |         actual_folder = subfolder.split("/")[-1]
 22 |         # print(actual_folder)
 23 |         if actual_folder.isdigit():
 24 |             fname = os.path.join(outfolder, "support_" + str(actual_folder) + ".txt")
 25 |             if os.path.isfile(fname):
 26 |                 g = open(fname, "r")
 27 |                 # read content from first file
 28 |                 for line in g:
 29 |                     # append content to second file
 30 |                     f.write(line)
 31 | def generate_low_abundance_mapping(outfolder):
 32 |     subfolders = [f.path for f in os.scandir(outfolder) if f.is_dir()]
 33 |     f = open(os.path.join(outfolder, "transcriptome_mapping_low.txt"), "w")
 34 |     for subfolder in subfolders:
 35 |         actual_folder = subfolder.split("/")[-1]
 36 |         # print(actual_folder)
 37 |         if actual_folder.isdigit():
 38 |             fname = os.path.join(outfolder, "cluster" + str(actual_folder) + "_mapping_low_abundance.txt")
 39 |             # print(fname)
 40 | 
 41 |             if os.path.isfile(fname):
 42 |                 g = open(fname, "r")
 43 |                 # read content from first file
 44 |                 for line in g:
 45 |                     # append content to second file
 46 |                     f.write(line)
 47 | def generate_single_mapping(outfolder):
 48 |     subfolders = [f.path for f in os.scandir(outfolder) if f.is_dir()]
 49 |     f = open(os.path.join(outfolder, "transcriptome_mapping.txt"), "w")
 50 |     for subfolder in subfolders:
 51 |         actual_folder = subfolder.split("/")[-1]
 52 |         if actual_folder.isdigit():
 53 |             fname = os.path.join(outfolder, "cluster" + str(actual_folder) + "_mapping.txt")
 54 | 
 55 |             if os.path.isfile(fname):
 56 |                 g = open(fname, "r")
 57 |                 # read content from first file
 58 |                 for line in g:
 59 |                     # append content to second file
 60 |                     f.write(line)
 61 | def generate_single_output(outfolder,write_fastq):
 62 |     subfolders = [f.path for f in os.scandir(outfolder) if f.is_dir()]
 63 |     if write_fastq:
 64 |         print("Generating transcriptome.fastq")
 65 |         f = open(os.path.join(outfolder,"transcriptome.fastq"), "w")
 66 |     else:
 67 |         print("Generating transcriptome.fasta")
 68 |         f = open(os.path.join(outfolder, "transcriptome.fasta"), "w")
 69 |     for subfolder in subfolders:
 70 |         #print("subfolder",subfolder)
 71 |         actual_folder=subfolder.split("/")[-1]
 72 |         #print(actual_folder)
 73 |         if actual_folder.isdigit():
 74 |             if write_fastq:
 75 |                 fname=os.path.join(outfolder, "cluster"+str(actual_folder)+"_merged.fq")
 76 |             else:
 77 |                 fname = os.path.join(outfolder, "cluster" + str(actual_folder) + "_merged.fa")
 78 |             #print(fname)
 79 |             if os.path.isfile(fname):
 80 |                 #print("True")
 81 |                 g = open(fname, "r")
 82 |                 # read content from first file
 83 |                 for line in g:
 84 |                     #if line.startswith('@'):
 85 |                     #    line=line+"_"+str(actual_folder)
 86 |                     #    print("LINE",line)
 87 |                     # append content to second file
 88 |                     f.write(line)
 89 |         
 90 | def generate_low_abundance_output(outfolder,write_fastq):
 91 |     subfolders = [f.path for f in os.scandir(outfolder) if f.is_dir()]
 92 |     if write_fastq:
 93 |         f = open(os.path.join(outfolder, "transcriptome_low.fastq"), "w")
 94 |     else:
 95 |         f = open(os.path.join(outfolder, "transcriptome_low.fasta"), "w")
 96 |     for subfolder in subfolders:
 97 |         actual_folder = subfolder.split("/")[-1]
 98 |                     # print(actual_folder)
 99 |         if actual_folder.isdigit():
100 |             if write_fastq:
101 |                 fname = os.path.join(outfolder, "cluster" + str(actual_folder) + "_merged_low_abundance.fq")
102 |                         # print(fname)
103 |             else:
104 |                 fname = os.path.join(outfolder, "cluster" + str(actual_folder) + "_merged_low_abundance.fa")
105 |             if os.path.isfile(fname):
106 |                 g = open(fname, "r")
107 |                 # read content from first file
108 |                 for line in g:
109 |                     if line.startswith('@') or line.startswith('>'):
110 |                         line = line + str(actual_folder)
111 |                     # append content to second file
112 |                     f.write(line)
113 | 
114 | 
115 | def remove_folders(outfolder):
116 |     subfolders = [f.path for f in os.scandir(outfolder) if f.is_dir()]
117 |     for subfolder in subfolders:
118 |         shutil.rmtree(os.path.join(outfolder,subfolder))
119 |             
120 | def generate_full_output(outfolder,write_fastq, write_low_abundance):
121 |     generate_single_output(outfolder, write_fastq)
122 |     generate_single_mapping(outfolder)
123 |     generate_single_support(outfolder)
124 |     if write_low_abundance:
125 |         generate_low_abundance_output(outfolder, write_fastq)
126 |         generate_low_abundance_mapping(outfolder)


--------------------------------------------------------------------------------
/isON_pipeline.sh:
--------------------------------------------------------------------------------
  1 | #!/bin/bash
  2 | set -e
  3 | #the pipeline can be run in different modes:
  4 | 
  5 | ####ONT data
  6 | # ont_with_pychopper: the full pipeline is run in addition to pychopper (pychopper, isONclust,isONcorrect,isONform)
  7 | # ont_no_pychopper: only the isONpipeline is run without pychopper (isONclust,isONcorrect, isONform)
  8 | # ont_no_pychop_isonclust3
  9 | ####PACBIO data
 10 | # pacbio: for PacBio data runs isONclust and isONform
 11 | 
 12 | 
 13 | ###### test modes (only for internal use)
 14 | # analysis: analysis of ont data:  isONclust,isONcorrect and isONform are run (e.g. analyses on the paper)
 15 | # only_isonform: only isONform is run
 16 | #!/bin/bash
 17 | 
 18 | programname=$0
 19 | function usage {
 20 |     echo ""
 21 |     echo "Runs the full isON pipeline. Please make sure that the input file has been preprocessed with pychopper for ONT data "
 22 |     echo ""
 23 |     echo "usage: $programname --raw_reads string --outfolder string --num_cores integer --isONform_folder string  --iso_abundance integer --mode string"
 24 |     echo ""
 25 |     echo "  --raw_reads   	        absolute path to the input file (in fastq format)"
 26 |     echo "                          (example: /home/user/Rawdata/raw_reads.fq)"
 27 |     echo "  --outfolder             absolute path to the output folder (the folder in which all outputs are stored)"
 28 |     echo "                          (example: /home/user/analysis_output)"
 29 |     echo "  --num_cores             the number of processors the pipeline may use"
 30 |     echo "                          (example: 8)"
 31 |     echo "  --isONform_folder       the absolute path to the isONform installation on your machine (leave empty if you have installed isONform via pip)"
 32 |     echo "                          (example: /home/user/isONform )"
 33 |     echo "  --iso_abundance         threshold which denotes the minimum read support neccessary for an isoform to be called (also minimum number of reads per cluster in isONclust)"
 34 |     echo "                          (example: 5)"
 35 |     echo "  --mode                  Run mode of the pipeline, possible modes are 'ont_no_pyc' and 'ont_with_pc' for ont data and 'pacbio' for pacbio data"
 36 |     echo "                          (example: ont_no_pychopper/ont_with_pychopper/pacbio)"
 37 |     echo " For ONT data: use 'ont_no_pychopper' if you want to run the isON pipeline and pychopper, use 'ont_with_pychopper' if you only want to run the isON pipeline. Please run pychopper yourself before running the pipeline."
 38 |     echo ""
 39 | }
 40 | 
 41 | while [ $# -gt 0 ]; do
 42 |     # Check if the current argument is "--help"
 43 |     if [[ $1 == "--help" ]]; then
 44 |         # Call the usage function and exit with status code 0
 45 |         usage
 46 |         exit 0
 47 |     # Check if the current argument is an option starting with "--"
 48 |     elif [[ $1 == "--"* ]]; then
 49 |         # Extract the option name by removing the leading dashes
 50 |         v="${1/--/}"
 51 |         # Check if the argument for this option was left empty (then we would have the next argument name as next entry)
 52 |         if [[ $2 != "--"* ]]; then
 53 |            #The argument was not left empty, therefore we properly set the argument as the value for option
 54 |            declare "$v"="$2"
 55 |            #We have to shift only in this case
 56 |            shift
 57 |         fi
 58 |     fi
 59 |     #This is the shift we have to perform each time
 60 |     shift
 61 | done
 62 | 
 63 | if [[ -z $raw_reads ]]; then
 64 |     usage
 65 |     die "Missing parameter --raw_reads"
 66 | elif [[ -z $outfolder ]]; then
 67 |     usage
 68 |     die "Missing parameter --outfolder"
 69 | elif [[ -z $mode ]]; then
 70 |     usage
 71 |     die "Missing parameter --mode"
 72 | #elif [[ -z $isONform_folder ]]; then
 73 | #    isONform_folder=''
 74 |     #TODO set isONform folder to '' if not given
 75 | fi
 76 | 
 77 | echo "Running `basename $0` raw reads: '$raw_reads' outfolder: '$outfolder' num_cores: '$num_cores' isONform_folder:'$isONform_folder' iso_abundance: '$iso_abundance' mode: '$mode'"
 78 | 
 79 | mkdir -p $outfolder
 80 | ##Testing whether the programs are installed properly before attempting to run them
 81 | # shellcheck disable=SC1072
 82 | if [ $mode != "pacbio" ] && [ $mode != "ont_no_pychop" ]; #we do not want to test for pychopper if we run modes pacbio or ont_no_pychop
 83 | then
 84 |   pychopper --h
 85 | fi
 86 | isONclust
 87 | #only test for isONcorrect if we want to use it (so not neccessary for Pacbio data)
 88 | if [ $mode != "pacbio" ]
 89 | then
 90 |   run_isoncorrect
 91 | fi
 92 | 
 93 | if [ -n "$isONform_folder"  ] #the user has given a path to isONform (cloned from github)
 94 |   then
 95 |   $isONform_folder/isONform_parallel --h
 96 | else
 97 |   python isONform_parallel --h
 98 | fi
 99 | 
100 | 
101 | 
102 | if [ $mode == "ont_with_pychopper" ]
103 | then
104 | echo
105 | echo "Will run pychopper (cdna_classifier.py), isONclust, isONcorrect and isONform. Make sure you have these tools installed."
106 | echo "For installation see: https://github.com/ksahlin/isONcorrect#installation and  https://github.com/aljpetri/isONform"
107 | echo
108 | 
109 | echo
110 | echo "Running pychopper"
111 | echo
112 | 
113 | pychopper  $raw_reads $outfolder/full_length.fq -t $num_cores
114 | 
115 | echo
116 | echo "Finished pychopper"
117 | echo
118 | 
119 | fi
120 | 
121 | if [ $mode != "only_isonform" ] # this if statement prevents isONclust and isONcorrect from being run
122 |   then
123 |     echo
124 |     echo "Running isONclust"
125 |     echo
126 |   if [ $mode == "ont_with_pychopper" ]
127 |     then
128 |         /usr/bin/time -v isONclust  --t $num_cores  --ont --fastq $outfolder/full_length.fq \
129 |              --outfolder $outfolder/clustering
130 |         /usr/bin/time -v isONclust write_fastq --N $iso_abundance --clusters $outfolder/clustering/final_clusters.tsv \
131 |                       --fastq $outfolder/full_length.fq --outfolder  $outfolder/clustering/fastq_files
132 |   elif [ $mode == "ont_no_pychopper" ]
133 |    then
134 |        /usr/bin/time -v  isONclust  --t $num_cores  --ont --fastq $raw_reads \
135 |              --outfolder $outfolder/clustering
136 |        /usr/bin/time -v isONclust write_fastq --N $iso_abundance --clusters $outfolder/clustering/final_clusters.tsv \
137 |                       --fastq $raw_reads --outfolder  $outfolder/clustering/fastq_files
138 |   elif [ $mode == "pacbio" ] #This is the pacbio mode
139 |     then
140 |        /usr/bin/time -v  isONclust  --t $num_cores  --isoseq  --fastq $raw_reads \
141 |              --outfolder $outfolder/clustering
142 |        /usr/bin/time -v isONclust write_fastq --N $iso_abundance --clusters $outfolder/clustering/final_clusters.tsv \
143 |                       --fastq $raw_reads --outfolder  $outfolder/clustering/fastq_files
144 | 
145 |   #elif [ $mode == "ont_no_pychop_isonclust3" ]
146 |   #  then
147 |   #    /home/alexanderpetri/Rust/isONclust_rs/target/release/isONclust3  --mode ont --fastq $raw_reads --outfolder $outfolder --n 1
148 |   else #[ $mode != "pacbio" ] && [ $mode != "'ont'" ]
149 |        /usr/bin/time -v  isONclust  --t $num_cores   --fastq $raw_reads \
150 |              --outfolder $outfolder
151 |        /usr/bin/time -v isONclust write_fastq --N $iso_abundance --clusters $outfolder/clustering/final_clusters.tsv \
152 |                       --fastq $raw_reads --outfolder  $outfolder/clustering/fastq_files
153 | 
154 | 
155 |   fi
156 | 
157 |   echo
158 |   echo "Finished isONclust"
159 |   echo
160 | #conda activate isON311
161 | 
162 |   if [ $mode != "pacbio" ]
163 |   then
164 |     echo
165 |     echo "Running isONcorrect"
166 |     echo
167 | 
168 |     /usr/bin/time -v  run_isoncorrect --t $num_cores  --fastq_folder $outfolder/clustering/fastq_files  --outfolder $outfolder/correction/
169 | 
170 |     echo
171 |     echo "Finished isONcorrect"
172 |     echo
173 |   fi
174 | fi
175 | echo
176 | echo "Merging reads back to single file. Corrected reads per cluster are still stored in: " $outfolder/correction/
177 | echo
178 | 
179 | echo
180 | echo "Running isONform"
181 | echo
182 | if [ -n "$isONform_folder"  ] #the user has given a path to isONform (cloned from github)
183 |   then
184 |   if [ $mode != "pacbio" ] #i.e. we run in ONT mode
185 |     then
186 |         /usr/bin/time -v  $isONform_folder/isONform_parallel --t $num_cores  --fastq_folder $outfolder/correction --exact_instance_limit 50 --k 20 --w 31 --xmin 14 --xmax 80 --max_seqs_to_spoa 200 --delta_len 10 --outfolder $outfolder/isoforms --iso_abundance $iso_abundance --delta_iso_len_3 30 --delta_iso_len_5 50
187 |   else #we run isONform in pacbio mode (adding the keyword clustered to the command)
188 |         /usr/bin/time -v   $isONform_folder/isONform_parallel --t $num_cores --fastq_folder $outfolder/clustering/fastq_files --exact_instance_limit 50 --k 20 --w 31 --xmin 14 --xmax 80 --max_seqs_to_spoa 200 --delta_len 10 --outfolder $outfolder/isoforms --iso_abundance $iso_abundance  --delta_iso_len_3 30 --delta_iso_len_5 50
189 |   fi
190 | 
191 | 
192 | else #the user has not given a path to isONform (pip installation supposed)
193 |   if [ $mode != "pacbio" ] #i.e. we run in ONT mode
194 |     then
195 |         /usr/bin/time -v  isONform_parallel --t $num_cores  --fastq_folder $outfolder/correction/ --exact_instance_limit 50 --k 20 --w 31 --xmin 14 --xmax 80 --max_seqs_to_spoa 200 --delta_len 10 --outfolder $outfolder/isoforms --iso_abundance $iso_abundance  --delta_iso_len_3 30 --delta_iso_len_5 50
196 |   else #we run isONform in pacbio mode (adding the keyword clustered to the command)
197 |         /usr/bin/time -v   isONform_parallel --t $num_cores --fastq_folder $outfolder/clustering/fastq_files --exact_instance_limit 50 --k 20 --w 31 --xmin 14 --xmax 80 --max_seqs_to_spoa 200 --delta_len 10 --outfolder $outfolder/isoforms --iso_abundance $iso_abundance --delta_iso_len_3 30 --delta_iso_len_5 50
198 |   fi
199 | fi
200 | echo
201 | echo "Finished isONform"
202 | echo
203 | 
204 | echo
205 | echo "Finished with pipeline and wrote corrected reads into: " $outfolder
206 | echo


--------------------------------------------------------------------------------
/modules/create_augmented_reference.py:
--------------------------------------------------------------------------------
  1 | from __future__ import print_function
  2 | import argparse
  3 | import os
  4 | import subprocess
  5 | from sys import stdout
  6 | from collections import defaultdict, deque
  7 | 
  8 | 
  9 | # Botonds DP
 10 | def cutoff(x):
 11 |     return x
 12 | 
 13 | """def longest_path_botond(dot_graph_path):
 14 |     G = nx.drawing.nx_pydot.read_dot(dot_graph_path)   
 15 |     topo = list(nx.topological_sort(G))   
 16 |     data = nx.get_node_attributes(G, 'label')
 17 |     # data = {x: chr(int(y.split(" - ")[1].split("\"")[0])) for x, y in data.items()}
 18 |     data = {x: y.split(" - ")[1].split("\"")[0] for x, y in data.items()}
 19 |     scores = defaultdict(int)
 20 |     bt = defaultdict(lambda: -1)
 21 |      
 22 |     for n in topo:
 23 |         sc = []
 24 |         for p in G.predecessors(n):
 25 |             tmp = G.get_edge_data(p, n, key=0)
 26 |             if 'label' in tmp:
 27 |                 weight = int(tmp['label'].strip("\""))
 28 |                 sc.append((p, scores[p] + cutoff(weight)))
 29 |                 # sc.append((p, scores[p] + weight))
 30 |         sc = sorted(sc, key=lambda x: x[1], reverse=True)
 31 |         if len(sc) < 1:
 32 |             continue
 33 |         bt[n] = sc[0][0]
 34 |         scores[n] = sc[0][1]
 35 |      
 36 |     max_score = max(scores.values())
 37 |     lookup = {y: x for x, y in scores.items()}
 38 |      
 39 |     prev = lookup[max_score]
 40 |     nodes = []
 41 |      
 42 |     while prev != -1:
 43 |         nodes.append(prev)
 44 |         prev = bt[prev]
 45 |      
 46 |     nodes = list(reversed(nodes))
 47 |      
 48 |     print(">cons_" + str(len(nodes)))
 49 |      
 50 |     bases = [data[n] for n in nodes]
 51 |     cons = ''.join(bases)
 52 |     print(cons)
 53 |     return cons"""
 54 | 
 55 | 
 56 | 
 57 | def run_spoa_affine(reads, ref_out_file, spoa_out_file, spoa_path, dot_graph_path):
 58 |     with open(spoa_out_file, "w") as output_file:
 59 |         print('Running spoa...', end=' ')
 60 |         stdout.flush()
 61 |         null = open("/dev/null", "w")
 62 |         subprocess.check_call([ spoa_path, "-q", reads, "-l", "2", "-r", "2", "-x", "-8", "-m", "10", "-o","-8", "-e", "-1", "-d", dot_graph_path], stderr=output_file, stdout=null)
 63 |         print('Done.')
 64 |         stdout.flush()
 65 |     output_file.close()
 66 |     l = open(spoa_out_file, "r").readlines()
 67 |     consensus = l[1].strip()
 68 |     msa = [s.strip() for s in l[3:]]
 69 |     print("affine heaviest path:", consensus, len(consensus))
 70 |     print()
 71 |     # print(msa)
 72 |     
 73 |     r = open(ref_out_file, "w")
 74 |     r.write(">{0}\n{1}".format("reference", consensus))
 75 |     r.close()
 76 |     return consensus, msa
 77 | 
 78 | def run_spoa_convex(reads, ref_out_file, spoa_out_file, spoa_path, dot_graph_path):
 79 |     with open(spoa_out_file, "w") as output_file:
 80 |         print('Running spoa convex...', end=' ')
 81 |         stdout.flush()
 82 |         null = open("/dev/null", "w")
 83 |         # print(spoa_path, "-l", "2", "-r", "2", "-g", "-4", "-e", "0", "-d", dot_graph_path, reads)
 84 |         subprocess.check_call([ spoa_path, "-l", "2", "-r", "2", "-m", "10", "-g", "-8", "-e", "-2", "-q", "-24", "-c", "-1" , "-d", dot_graph_path, reads], stdout=output_file, stderr=null)
 85 |         # subprocess.check_call([ spoa_path, "-q", reads, "-l", "2", "-r", "2", "-o","-4", "-e", "0", "-d", dot_graph_path], stderr=output_file, stdout=null)
 86 |         print('Done.')
 87 |         stdout.flush()
 88 |     output_file.close()
 89 |     l = open(spoa_out_file, "r").readlines()
 90 |     consensus = l[1].strip()
 91 |     msa = [s.strip() for s in l[3:]]
 92 |     print("convex heaviest path:", consensus, len(consensus))
 93 |     print()
 94 |     # print(msa)
 95 |     
 96 |     r = open(ref_out_file, "w")
 97 |     r.write(">{0}\n{1}".format("reference", consensus))
 98 |     r.close()
 99 |     return consensus, msa
100 | 
101 | def run_spoa_affine_v2_0_3(reads, ref_out_file, spoa_out_file, spoa_path, dot_graph_path):
102 |     with open(spoa_out_file, "w") as output_file:
103 |         print('Running spoa...', end=' ')
104 |         stdout.flush()
105 |         null = open("/dev/null", "w")
106 |         # print(spoa_path, "-l", "2", "-r", "2", "-g", "-4", "-e", "0", "-d", dot_graph_path, reads)
107 |         subprocess.check_call([ spoa_path, "-l", "2", "-r", "2", "-x", "-4", "-m", "10", "-g", "-8", "-e", "-1", "-d", dot_graph_path, reads], stdout=output_file, stderr=null)
108 |         # subprocess.check_call([ spoa_path, "-q", reads, "-l", "2", "-r", "2", "-o","-4", "-e", "0", "-d", dot_graph_path], stderr=output_file, stdout=null)
109 |         print('Done.')
110 |         stdout.flush()
111 |     output_file.close()
112 |     l = open(spoa_out_file, "r").readlines()
113 |     consensus = l[1].strip()
114 |     msa = [s.strip() for s in l[3:]]
115 |     print("affine_v2_0_2 heaviest path:", consensus, len(consensus))
116 |     print()
117 |     # print(msa)
118 |     
119 |     r = open(ref_out_file, "w")
120 |     r.write(">{0}\n{1}".format("reference", consensus))
121 |     r.close()
122 |     return consensus, msa
123 | 
124 | def run_spoa_with_msa(reads, ref_out_file, spoa_out_file, spoa_path, dot_graph_path):
125 |     with open(spoa_out_file, "w") as output_file:
126 |         # print('Running spoa...', end=' ')
127 |         stdout.flush()
128 |         null = open("/dev/null", "w")
129 |         # subprocess.check_call([ spoa_path, reads, "-l", "0", "-r", "2", "-g", "-2", "-d", dot_graph_path], stdout=output_file, stderr=null)
130 |         subprocess.check_call([ spoa_path, reads, "-l", "0", "-r", "2", "-g", "-2"], stdout=output_file, stderr=null)
131 |         # print('Done.')
132 |         stdout.flush()
133 |     output_file.close()
134 |     l = open(spoa_out_file, "r").readlines()
135 |     consensus = l[1].strip()
136 |     msa = [s.strip() for s in l[3:]]
137 |     # print("regular spoa:", consensus)
138 |     # print(len(consensus))
139 |     # print(msa)
140 |     
141 |     r = open(ref_out_file, "w")
142 |     r.write(">{0}\n{1}".format("reference", consensus))
143 |     r.close()
144 | 
145 |     return consensus, msa
146 | 
147 | def run_spoa(reads, spoa_out_file, spoa_path):
148 |     with open(spoa_out_file, "w") as output_file:
149 |         # print('Running spoa...', end=' ')
150 |         stdout.flush()
151 |         null = open("/dev/null", "w")
152 |         subprocess.check_call([ spoa_path, reads, "-l", "0", "-r", "0", "-g", "-2"], stdout=output_file, stderr=null)
153 |         # print('Done.')
154 |         stdout.flush()
155 |     # output_file.close()
156 |     l = open(spoa_out_file, "r").readlines()
157 |     consensus = l[1].strip()
158 |     del l
159 |     return consensus
160 | 
161 | def run_spoa_m(reads, spoa_out_file, spoa_path):
162 |     with open(spoa_out_file, "w") as output_file:
163 |         # print('Running spoa...', end=' ')
164 |         stdout.flush()
165 |         null = open("/dev/null", "w")
166 |         subprocess.check_call([ spoa_path, reads, "-l", "0", "-r", "0", "-g", "-4"], stdout=output_file, stderr=null)
167 |         # print('Done.')
168 |         stdout.flush()
169 |     # output_file.close()
170 |     l = open(spoa_out_file, "r").readlines()
171 |     consensus = l[1].strip()
172 |     del l
173 |     return consensus
174 | 
175 | def run_spoa_m2(reads, spoa_out_file, spoa_path):
176 |     with open(spoa_out_file, "w") as output_file:
177 |         # print('Running spoa...', end=' ')
178 |         stdout.flush()
179 |         null = open("/dev/null", "w")
180 |         subprocess.check_call([ spoa_path, reads, "-l", "0", "-r", "0", "-g", "-8"], stdout=output_file, stderr=null)
181 |         # print('Done.')
182 |         stdout.flush()
183 |     # output_file.close()
184 |     l = open(spoa_out_file, "r").readlines()
185 |     consensus = l[1].strip()
186 |     del l
187 |     return consensus
188 | 
189 | def run_racon(reads_to_center, read_alignments_paf, center_file, outfolder, cores, racon_iter):
190 |     racon_stdout = os.path.join(outfolder, "racon_stdout.txt")
191 | 
192 | 
193 |     with open(racon_stdout, "w") as output_file:
194 |         # print('Running medaka...', end=' ')
195 |         stdout.flush()
196 |         for i in range(racon_iter):
197 |             # read_alignments = open(os.path.join(outfolder, "read_alignments_it_{0}.paf".format(i)), 'w')
198 |             # mm2_stderr = open(os.path.join(outfolder, "mm2_stderr_it_{0}.txt".format(i)), "w")
199 |             racon_stderr = open(os.path.join(outfolder, "racon_stderr_it_{0}.txt".format(i)), "w")
200 |             racon_polished = open(os.path.join(outfolder, "racon_polished_it_{0}.fasta".format(i)), 'w')
201 | 
202 |             # subprocess.check_call(['minimap2', '-k 9', '-w 1', '-f 0.00001', '-n 2', '-m 10', center_file, reads_to_center], stdout=read_alignments, stderr=mm2_stderr)
203 |             subprocess.check_call(['racon', reads_to_center, read_alignments_paf, center_file], stdout=racon_polished, stderr=racon_stderr)
204 |             center_file = racon_polished.name
205 |     l = open(center_file, "r").readlines()
206 |     if len(l) == 2:
207 |         consensus = l[1].strip()
208 |     else:
209 |         consensus = ''
210 |     del l
211 |     return consensus
212 | 
213 | # For eventual De Bruijn graph approach
214 | 
215 | def kmer_counter(reads, k_size):
216 |     count = defaultdict(int)
217 |     position_count = defaultdict(list)
218 | 
219 |     for r_i in reads:
220 |         (acc, seq, qual) = reads[r_i]
221 |         # seq_hpol_comp = ''.join(ch for ch, _ in itertools.groupby(seq))
222 |         read_kmers = deque([seq[i:i+k_size] for i in range(len(seq) - k_size )])
223 |         for i, kmer in enumerate(read_kmers):
224 |             count[kmer] += 1
225 | 
226 |             position_count[kmer].append( (r_i, i))
227 | 
228 |     # cnt_sorted = sorted(count.items(), key = lambda x: x[1], reverse=True)
229 |     # print(len(cnt_sorted),cnt_sorted)
230 |     return count, position_count
231 | 
232 | 
233 | if __name__ == '__main__':
234 |     parser = argparse.ArgumentParser(description="Maps the given reads with bwa.")
235 |     parser.add_argument('reads', type=str, help='Fasta or fastq')
236 |     parser.add_argument('outfile', type=str, help='Fasta or fastq')
237 |     parser.add_argument('--spoa_path', type=str, default='spoa', required=False, help='Path to spoa binary with bwa binary name at the end.')
238 | 
239 | 
240 |     args = parser.parse_args()
241 | 
242 |     run_spoa(args.reads, args.outfile, args.spoa_path)
243 | 


--------------------------------------------------------------------------------
/modules/consensus.py:
--------------------------------------------------------------------------------
  1 | from __future__ import print_function
  2 | import subprocess
  3 | import sys, os
  4 | from sys import stdout
  5 | import re 
  6 | import shutil
  7 | import parasail
  8 | import glob
  9 | 
 10 | from modules import help_functions
 11 | 
 12 | 
 13 | def cigar_to_seq(cigar, query, ref):
 14 |     cigar_tuples = []
 15 |     result = re.split(r'[=DXSMI]+', cigar)
 16 |     cig_pos = 0
 17 |     for length in result[:-1]:
 18 |         cig_pos += len(length)
 19 |         type_ = cigar[cig_pos]
 20 |         cig_pos += 1
 21 |         cigar_tuples.append((int(length), type_ ))
 22 | 
 23 |     r_index = 0
 24 |     q_index = 0
 25 |     q_aln = []
 26 |     r_aln = []
 27 |     for length_ , type_ in cigar_tuples:
 28 |         if type_ == "=" or type_ == "X":
 29 |             q_aln.append(query[q_index : q_index + length_])
 30 |             r_aln.append(ref[r_index : r_index + length_])
 31 | 
 32 |             r_index += length_
 33 |             q_index += length_
 34 |         
 35 |         elif  type_ == "I":
 36 |             # insertion w.r.t. reference
 37 |             r_aln.append('-' * length_)
 38 |             q_aln.append(query[q_index: q_index + length_])
 39 |             #  only query index change
 40 |             q_index += length_
 41 | 
 42 |         elif type_ == 'D':
 43 |             # deletion w.r.t. reference
 44 |             r_aln.append(ref[r_index: r_index + length_])
 45 |             q_aln.append('-' * length_)
 46 |             #  only ref index change
 47 |             r_index += length_
 48 |         
 49 |         else:
 50 |             sys.exit()
 51 | 
 52 |     return "".join([s for s in q_aln]), "".join([s for s in r_aln]), cigar_tuples
 53 | 
 54 | 
 55 | def parasail_alignment(s1, s2, match_score = 2, mismatch_penalty = -2, opening_penalty = 12, gap_ext = 1):
 56 |     user_matrix = parasail.matrix_create("ACGT", match_score, mismatch_penalty)
 57 |     result = parasail.sg_trace_scan_16(s1, s2, opening_penalty, gap_ext, user_matrix)
 58 |     if result.saturated:
 59 |         result = parasail.sg_trace_scan_32(s1, s2, opening_penalty, gap_ext, user_matrix)
 60 | 
 61 |     # difference in how to obtain string from parasail between python v2 and v3... 
 62 |     if sys.version_info[0] < 3:
 63 |         cigar_string = str(result.cigar.decode).decode('utf-8')
 64 |     else:
 65 |         cigar_string = str(result.cigar.decode, 'utf-8')
 66 |     s1_alignment, s2_alignment, cigar_tuples = cigar_to_seq(cigar_string, s1, s2)
 67 |     return s1_alignment, s2_alignment, cigar_string, cigar_tuples, result.score
 68 | 
 69 | 
 70 | def reverse_complement(string):
 71 |     # Modified for Abyss output
 72 |     rev_nuc = {'A':'T', 'C':'G', 'G':'C', 'T':'A', 'a':'t', 'c':'g', 'g':'c', 't':'a', 'N':'N', 'X':'X', 'n':'n', 'Y':'R', 'R':'Y', 'K':'M', 'M':'K', 'S':'S', 'W':'W', 'B':'V', 'V':'B', 'H':'D', 'D':'H', 'y':'r', 'r':'y', 'k':'m', 'm':'k', 's':'s', 'w':'w', 'b':'v', 'v':'b', 'h':'d', 'd':'h'}
 73 |     rev_comp = ''.join([rev_nuc[nucl] for nucl in reversed(string)])
 74 |     return rev_comp
 75 | 
 76 | 
 77 | def run_spoa(reads, spoa_out_file, spoa_path):
 78 |     with open(spoa_out_file, "w") as output_file:
 79 |         # print('Running spoa...', end=' ')
 80 |         stdout.flush()
 81 |         null = open("/dev/null", "w")
 82 |         subprocess.check_call([ spoa_path, reads, "-l", "0", "-r", "0", "-g", "-2"], stdout=output_file, stderr=null)
 83 |         # print('Done.')
 84 |         stdout.flush()
 85 |     output_file.close()
 86 |     l = open(spoa_out_file, "r").readlines()
 87 |     consensus = l[1].strip()
 88 |     return consensus
 89 | 
 90 | 
 91 | def run_medaka(reads_to_center, center_file, outfolder, cores, medaka_model):
 92 |     medaka_stdout = os.path.join(outfolder, "stdout.txt")
 93 |     with open(medaka_stdout, "w") as output_file:
 94 |         # print('Running medaka...', end=' ')
 95 |         stdout.flush()
 96 |         medaka_stderr = open(os.path.join(outfolder, "stderr.txt"), "w")
 97 |         if medaka_model:
 98 |             subprocess.check_call(['medaka_consensus', '-i', reads_to_center, "-d", center_file, "-o", outfolder, "-t", cores, "-m", medaka_model], stdout=output_file, stderr=medaka_stderr)
 99 |         else:
100 |             subprocess.check_call(['medaka_consensus', '-i', reads_to_center, "-d", center_file, "-o", outfolder, "-t", cores], stdout=output_file, stderr=medaka_stderr)
101 |         stdout.flush()
102 |     output_file.close()
103 | 
104 | 
105 | def run_racon(reads_to_center, center_file, outfolder, cores, racon_iter):
106 |     racon_stdout = os.path.join(outfolder, "stdout.txt")
107 |     with open(racon_stdout, "w") as output_file:
108 |         # print('Running medaka...', end=' ')
109 |         stdout.flush()
110 |         for i in range(racon_iter):
111 |             read_alignments = open(os.path.join(outfolder, "read_alignments_it_{0}.paf".format(i)), 'w')
112 |             mm2_stderr = open(os.path.join(outfolder, "mm2_stderr_it_{0}.txt".format(i)), "w")
113 |             racon_stderr = open(os.path.join(outfolder, "racon_stderr_it_{0}.txt".format(i)), "w")
114 |             racon_polished = open(os.path.join(outfolder, "racon_polished_it_{0}.fasta".format(i)), 'w')
115 | 
116 |             subprocess.check_call(['minimap2', '-x', 'map-ont', center_file, reads_to_center], stdout=read_alignments, stderr=mm2_stderr)
117 |             subprocess.check_call(['racon', reads_to_center, read_alignments.name, center_file], stdout=racon_polished, stderr=racon_stderr)
118 |             center_file = racon_polished.name
119 | 
120 |         shutil.copyfile(center_file, os.path.join(outfolder, "consensus.fasta"))
121 |         # print('Done.')
122 |         stdout.flush()
123 |     output_file.close()
124 | 
125 | 
126 | def detect_reverse_complements(centers, rc_identity_threshold):
127 |     filtered_centers = []
128 |     already_removed = set()
129 |     for i, (nr_reads_in_cl, c_id, seq, reads_path) in enumerate(centers):
130 |         if type(reads_path) != list:
131 |             all_reads = [reads_path]
132 |         else:
133 |             all_reads = reads_path
134 | 
135 |         merged_cluster_id = c_id
136 |         merged_nr_reads = nr_reads_in_cl
137 |         if c_id in already_removed:
138 |             #print("has already been merged, skipping")
139 |             continue
140 | 
141 |         elif i == len(centers) - 1: # last sequence and it is not in already removed
142 |             filtered_centers.append( [merged_nr_reads, c_id, seq, all_reads ] )
143 | 
144 |         else:
145 |             for j, (nr_reads_in_cl2, c_id2, seq2, reads_path) in enumerate(centers[i+1:]):
146 |                 seq2_rc = reverse_complement(seq2)
147 |                 seq_aln, seq2_rc_aln, cigar_string, cigar_tuples, alignment_score = parasail_alignment(seq, seq2_rc)
148 |                 # print(i,j)
149 |                 # print(seq_aln)
150 |                 # print(seq2_rc_aln)
151 |                 nr_mismatching_pos = len([1 for n1, n2 in zip(seq_aln, seq2_rc_aln) if n1 != n2])
152 |                 total_pos = len(seq_aln)
153 |                 aln_identity =  (total_pos - nr_mismatching_pos) / float(total_pos)
154 |                 #print(aln_identity)
155 | 
156 |                 if aln_identity >= rc_identity_threshold:
157 |                     #print("Detected alignment identidy above threchold for reverse complement. Keeping center with the most read support and adding rc reads to supporting reads.")
158 |                     merged_nr_reads += nr_reads_in_cl2
159 |                     already_removed.add(c_id2)
160 | 
161 |                     if type(reads_path) != list:
162 |                         all_reads.append(reads_path)
163 |                     else:
164 |                         for rp in reads_path:
165 |                             all_reads.append(rp)
166 | 
167 |             filtered_centers.append( [merged_nr_reads, c_id, seq, all_reads] )
168 | 
169 |     #print(len(filtered_centers), "consensus formed.")
170 |     return filtered_centers
171 | 
172 | 
173 | def polish_sequences(centers, args):
174 |     print("Saving spoa references to files:", os.path.join(args.outfolder, "consensus_reference_X.fasta"))
175 |     # printing output from spoa and grouping reads
176 |     # to_polishing = []
177 |     if args.medaka:
178 |         polishing_pattern = os.path.join(args.outfolder, "medaka_cl_id_*")
179 |     elif args.racon:
180 |         polishing_pattern = os.path.join(args.outfolder, "racon_cl_id_*")
181 | 
182 |     for folder in glob.glob(polishing_pattern):
183 |         shutil.rmtree(folder)
184 | 
185 |     spoa_pattern = os.path.join(args.outfolder, "consensus_reference_*")
186 |     for file in glob.glob(spoa_pattern):
187 |         os.remove(file)
188 | 
189 |     for i, (nr_reads_in_cluster, c_id, center, all_reads) in enumerate(centers):
190 |         # print('lol',c_id,center)
191 |         spoa_center_file = os.path.join(args.outfolder, "consensus_reference_{0}.fasta".format(c_id))
192 |         f = open(spoa_center_file, "w")
193 |         f.write(">{0}\n{1}\n".format("consensus_cl_id_{0}_total_supporting_reads_{1}".format(c_id, nr_reads_in_cluster), center))
194 |         f.close()
195 |         
196 |         all_reads_file = os.path.join(args.outfolder, "reads_to_consensus_{0}.fastq".format(c_id))
197 |         f = open(all_reads_file, "w")
198 |         for fasta_file in all_reads: 
199 |             reads = { acc : (seq, qual) for acc, (seq, qual) in help_functions.readfq(open(fasta_file, 'r'))}
200 |             for acc, (seq, qual) in reads.items():
201 |                 f.write("@{0}\n{1}\n{2}\n{3}\n".format(acc, seq, "+", qual))
202 |         f.close()
203 |         # to_polishing.append( (nr_reads_in_cluster, c_id, spoa_center_file, all_reads_file) )
204 | 
205 |         if args.medaka:
206 |             print("running medaka on spoa reference {0}.".format(c_id))
207 |             # for (nr_reads_in_cluster, c_id, spoa_center_file, all_reads_file) in to_polishing:
208 |             polishing_outfolder = os.path.join(args.outfolder, "medaka_cl_id_{0}".format(c_id))
209 |             help_functions.mkdir_p(polishing_outfolder)
210 |             run_medaka(all_reads_file, spoa_center_file, polishing_outfolder, "1", args.medaka_model)
211 |             print("Saving medaka reference to file:", os.path.join(args.outfolder, "medaka_cl_id_{0}/consensus.fasta".format(c_id)))   
212 |             l = open(os.path.join(polishing_outfolder, "consensus.fasta"), 'r').readlines()
213 |             center_polished = l[1].strip()
214 |             centers[i][2] = center_polished
215 |         elif args.racon:
216 |             print("running racon on spoa reference {0}.".format(c_id))
217 |             # for (nr_reads_in_cluster, c_id, spoa_center_file, all_reads_file) in to_polishing:
218 |             polishing_outfolder = os.path.join(args.outfolder, "racon_cl_id_{0}".format(c_id))
219 |             help_functions.mkdir_p(polishing_outfolder)
220 |             run_racon(all_reads_file, spoa_center_file, polishing_outfolder, "1", args.racon_iter)
221 |             print("Saving racon reference to file:", os.path.join(args.outfolder, "racon_cl_id_{0}/consensus.fasta".format(c_id)))   
222 |             l = open(os.path.join(polishing_outfolder, "consensus.fasta"), 'r').readlines()
223 |             center_polished = l[1].strip()
224 |             centers[i][2] = center_polished
225 | 
226 |     f.close()
227 |     return centers
228 | 
229 | 
230 | def form_draft_consensus(clusters, representatives, sorted_reads_fastq_file, work_dir, abundance_cutoff, args):
231 |     centers = []
232 |     reads = { acc : (seq, qual) for acc, (seq, qual) in help_functions.readfq(open(sorted_reads_fastq_file, 'r'))}
233 |     for c_id, all_read_acc in sorted(clusters.items(), key = lambda x: (len(x[1]),representatives[x[0]][5]), reverse=True):
234 |         reads_path = open(os.path.join(work_dir, "reads_c_id_{0}.fq".format(c_id)), "w")
235 |         nr_reads_in_cluster = len(all_read_acc)
236 |         # print("nr_reads_in_cluster", nr_reads_in_cluster)
237 |         if nr_reads_in_cluster >= abundance_cutoff:
238 |             for acc in all_read_acc:
239 |                 seq, qual = reads[acc]
240 |                 reads_path.write("@{0}\n{1}\n{2}\n{3}\n".format(acc, seq, "+", qual))
241 |                 # reads_path.write(">{0}\n{1}\n".format(str(q_id)+str(pos1)+str(pos2), seq))
242 |             reads_path.close()
243 |             # spoa_ref = create_augmented_reference.run_spoa(reads_path.name, os.path.join(work_dir,"spoa_tmp.fa"), "spoa")
244 |             #print("creating center of {0} sequences.".format(nr_reads_in_cluster))
245 |             center = run_spoa(reads_path.name, os.path.join(work_dir,"spoa_tmp.fa"), "spoa")
246 |             centers.append( [nr_reads_in_cluster, c_id, center, reads_path.name])
247 |     return centers
248 | 


--------------------------------------------------------------------------------
/modules/batch_merging_parallel.py:
--------------------------------------------------------------------------------
  1 | import itertools
  2 | import pickle
  3 | import os
  4 | 
  5 | import shutil
  6 | from modules import IsoformGeneration
  7 | from modules import Parallelization_side_functions
  8 | 
  9 | class Read:
 10 |     def __init__(self, sequence, reads, merged):
 11 |         self.sequence = sequence
 12 |         self.reads = reads
 13 |         self.merged = merged
 14 | 
 15 | def generate_consensus_path(work_dir, mappings1, mappings2, all_sequences, spoa_count):
 16 |     """ This method is used to generate the consensus file needed for the consensus generation
 17 |     INPUT:  work_dir  : The working directory in which to store the file
 18 |     OUTPUT: spoa_ref:   The consensus
 19 |     """
 20 |     reads_path = open(os.path.join(work_dir, "reads_tmp.fa"), "w")
 21 |     seq_count = 0
 22 |     for id in mappings1:
 23 |         if seq_count < spoa_count:
 24 |             sequence = all_sequences[id]
 25 |             reads_path.write(">{0}\n{1}\n".format(str(id), sequence))
 26 |             seq_count += 1
 27 |     for id in mappings2:
 28 |         if seq_count < spoa_count:
 29 |             if not (id in all_sequences):
 30 |                 sequence = id
 31 |                 reads_path.write(">{0}\n{1}\n".format(str(id), sequence))
 32 |                 seq_count += 1
 33 |     reads_path.close()
 34 |     spoa_ref = IsoformGeneration.run_spoa(reads_path.name, os.path.join(work_dir, "spoa_tmp.fa"))
 35 |     return spoa_ref
 36 | 
 37 | 
 38 | """ This function reads the consensus file and saves the infos in batch_reads_id
 39 | INPUT:  work_dir  : The working directory in which to store the file
 40 | OUTPUT: spoa_ref:   The consensus
 41 | """
 42 | 
 43 | 
 44 | def read_spoa_file(batch_id, cl_dir):
 45 |     filename = "spoa" + str(batch_id) + "merged.fasta"
 46 |     batch_reads_id = {}
 47 |     with open(os.path.join(cl_dir, filename)) as f:
 48 |         for id, sequence in itertools.zip_longest(*[f] * 2):
 49 |             inter_id = id.replace('\n', '')
 50 |             new_id = int(inter_id.replace('>consensus', ''))
 51 |             sequence = sequence.replace('\n', '')
 52 |             batch_reads_id[new_id] = sequence
 53 |     return batch_reads_id
 54 | 
 55 | 
 56 | def read_mapping_file(batch_id, cl_dir):
 57 |     batch_mappings_id = {}
 58 |     mappingname = "mapping" + str(batch_id) + ".txt"
 59 |     incoming_ctr = 0
 60 |     with open(os.path.join(cl_dir, mappingname)) as g:
 61 |         for id, reads in itertools.zip_longest(*[g] * 2):
 62 |             inter_id = id.replace('\n', '')
 63 |             id = int(inter_id.replace('consensus', ''))
 64 |             reads = reads.replace('\n', '')
 65 |             reads = reads.replace('[', '')
 66 |             reads = reads.replace(']', '')
 67 |             reads = reads.replace("'", "")
 68 |             readslist = reads.split(", ")
 69 |             batch_mappings_id[id] = readslist
 70 |             incoming_ctr += len(readslist)
 71 |     return batch_mappings_id
 72 | 
 73 | 
 74 | def read_batch_file(batch_id, all_infos_dict, all_reads_dict, cl_dir):
 75 |     batchfilename = str(batch_id) + "_batchfile.fa"
 76 |     with open(os.path.join(cl_dir, batchfilename)) as h:
 77 |         for id, seq in itertools.zip_longest(*[h] * 2):
 78 |             id = id.replace('\n', '')
 79 |             id = id.replace('>', '')
 80 |             seq = seq.replace('\n', '')
 81 |             all_reads_dict[id] = seq
 82 |             all_infos_dict[batch_id] = {}
 83 | 
 84 | 
 85 | def write_final_output(all_infos_dict, outfolder, iso_abundance, cl_dir, folder, write_fastq, write_low_abundance):
 86 |     #write_low_abundance = False
 87 |     support_name = "support_" + str(folder) + ".txt"
 88 |     other_support_name = "support_" + str(folder) + "low_abundance.txt"
 89 |     other_mapping_name = "cluster" + str(folder) + "_mapping_low_abundance.txt"
 90 |     if write_fastq:
 91 |         consensus_name = "cluster" + str(folder) + "_merged.fq"
 92 |         if write_low_abundance:
 93 |             other_consensus_name = "cluster" + str(folder) + "_merged_low_abundance.fq"
 94 |             other_consensus = open(os.path.join(outfolder, other_consensus_name), 'w')
 95 |             other_mapping = open(os.path.join(outfolder, other_mapping_name), 'w')
 96 |             other_support_file = open(os.path.join(outfolder, other_support_name), "w")
 97 |     else:
 98 |         consensus_name = "cluster" + str(folder) + "_merged.fa"
 99 |         if write_low_abundance:
100 |             other_consensus_name = "cluster" + str(folder) + "_merged_low_abundance.fa"
101 |             other_consensus = open(os.path.join(outfolder, other_consensus_name), 'w')
102 |             other_mapping = open(os.path.join(outfolder, other_mapping_name), 'w')
103 |             other_support_file = open(os.path.join(outfolder, other_support_name), "w")
104 | 
105 |     mapping_name = "cluster" + str(folder) + "_mapping.txt"
106 |     support_file = open(os.path.join(outfolder, support_name), "w")
107 |     consensus_file = open(os.path.join(outfolder, consensus_name), "w")
108 |     mapping_file = open(os.path.join(outfolder, mapping_name), 'w')
109 |     skipped_reads = {}
110 |     for batchid, id_dict in all_infos_dict.items():
111 |         for id, infos in id_dict.items():
112 |             new_id = str(folder) + "_" + str(batchid) + "_" + str(id)
113 |             if not infos.merged:
114 |                 if len(infos.reads) >= iso_abundance or iso_abundance == 1:
115 |                     mapping_file.write(">{0}\n{1}\n".format(new_id, all_infos_dict[batchid][id].reads))
116 |                     if write_fastq:
117 |                         consensus_file.write("@{0}\n{1}\n+\n{2}\n".format(new_id, all_infos_dict[batchid][id].sequence,
118 |                                                                       "+" * len(all_infos_dict[batchid][id].sequence)))
119 |                     else:
120 |                         consensus_file.write(">{0}\n{1}\n".format(new_id, all_infos_dict[batchid][id].sequence))
121 |                     support_file.write("{0}: {1}\n".format(new_id, len(all_infos_dict[batchid][id].reads)))
122 |                 else:
123 |                     if write_low_abundance:
124 |                         if write_fastq:
125 |                             other_consensus.write("@{0}\n{1}\n+\n{2}\n".format(new_id, all_infos_dict[batchid][id].sequence,
126 |                                                                            "+" * len(
127 |                                                                                all_infos_dict[batchid][id].sequence)))
128 |                         else:
129 |                             other_consensus.write(">{0}\n{1}\n".format(new_id, all_infos_dict[batchid][id].sequence))
130 |                         if new_id in all_infos_dict:
131 |                             other_support_file.write("{0}: {1}\n".format(new_id, len(all_infos_dict[new_id].reads)))
132 |                         else:
133 |                             other_support_file.write("{0}: {1}\n".format(new_id, 1))
134 |                         other_mapping.write(">{0}\n{1}\n".format(new_id, all_infos_dict[batchid][id].reads))
135 |     if write_low_abundance:
136 |         for skipfile in os.listdir(cl_dir):
137 |             if skipfile.startswith("skip"):
138 |                 with open(os.path.join(cl_dir, skipfile)) as h:
139 |                     for id, seq in itertools.zip_longest(*[h] * 2):
140 |                         id = id.replace('\n', '')
141 |                         id = id.replace('>', '')
142 |                         seq = seq.replace('\n', '')
143 |                         skipped_reads[id] = seq
144 | 
145 |                 for acc, seq in skipped_reads.items():
146 |                     other_consensus.write("@{0}\n{1}\n+\n{2}\n".format(acc, seq, "+" * len(seq)))
147 |                     other_support_file.write("{0}: {1}\n".format(acc, len(seq)))
148 |                     other_mapping.write(">{0}\n{1}\n".format(acc, acc))
149 | 
150 |     consensus_file.close()
151 |     mapping_file.close()
152 |     if write_low_abundance:
153 |         other_consensus.close()
154 |         other_mapping.close()
155 | 
156 | 
157 | # TODO: add the rest of variables for this method and move filewriting to wrappermethod
158 | 
159 | 
160 | def actual_merging_process(all_infos_dict, delta, delta_len,
161 |                            delta_iso_len_3, delta_iso_len_5, max_seqs_to_spoa, all_batch_sequences, work_dir):
162 |     all_infos_list = sorted(all_infos_dict.items(), key=lambda x: x[0], reverse=True)
163 |     for b_i, (batchid, id_dict) in enumerate(all_infos_list[:len(all_infos_list) - 1]):
164 |         batch_id_list = sorted(id_dict.items(), key=lambda x: len(x[1].sequence))
165 |         for b_j, (batchid2, id_dict2) in enumerate(all_infos_list[b_i + 1:]):
166 |             batch_id_list2 = sorted(id_dict2.items(), key=lambda x: len(x[1].sequence))
167 |             if not batchid2 <= batchid:
168 |                 for t_id1, (id, infos) in enumerate(batch_id_list):
169 |                     if not infos.merged:
170 |                         for t_id2, (id2, infos2) in enumerate(batch_id_list2):
171 |                             if len(infos2.sequence) >= len(infos.sequence):
172 |                                 batch_id_long = batchid2
173 |                                 batch_id_short = batchid
174 |                                 id_long = id2
175 |                                 infos_long = infos2
176 |                                 id_short = id
177 |                                 infos_short = infos
178 |                             else:
179 |                                 batch_id_long = batchid
180 |                                 batch_id_short = batchid2
181 |                                 id_long = id
182 |                                 infos_long = infos
183 |                                 id_short = id2
184 |                                 infos_short = infos2
185 | 
186 |                             if infos_long.merged:
187 |                                 continue
188 |                             if DEBUG:
189 |                                 print("bid", batchid, ": ", id, "bid2", batchid2, ": ", id2)
190 |                             if not infos_short.merged:
191 |                                 consensus1 = infos_long.sequence
192 |                                 consensus2 = infos_short.sequence
193 |                                 good_to_pop = IsoformGeneration.align_to_merge(consensus1, consensus2, delta, delta_len,
194 |                                                                                 delta_iso_len_3,
195 |                                                                                delta_iso_len_5)
196 |                                 if DEBUG:
197 |                                     print(good_to_pop)
198 |                                 if good_to_pop:
199 |                                     # if the first combi has more than 50 reads support
200 |                                     if len(infos_long.reads) > 50:
201 |                                         all_infos_dict[batch_id_long][
202 |                                             id_long].reads = infos_long.reads + infos_short.reads
203 |                                         infos_short.merged = True
204 |                                     else:
205 |                                         mappings1 = infos_long.reads
206 |                                         mappings2 = infos_short.reads
207 |                                         new_consensus = generate_consensus_path(work_dir, mappings1, mappings2,
208 |                                                                                 all_batch_sequences, max_seqs_to_spoa)
209 |                                         all_infos_dict[batch_id_long][id_long].sequence = new_consensus
210 |                                         all_infos_dict[batch_id_long][
211 |                                             id_long].reads = infos_long.reads + infos_short.reads
212 |                                         all_infos_dict[batch_id_short][id_short].merged = True
213 | 
214 | 
215 | def join_back_via_batch_merging(outdir, delta, delta_len, delta_iso_len_3,
216 |                                 delta_iso_len_5, max_seqs_to_spoa, iso_abundance,write_fastq,write_low_abundance):
217 |     print("Batch Merging")
218 |     unique_cl_ids = set()
219 |     subfolders = [f.path for f in os.scandir(outdir) if f.is_dir()]
220 |     # iterate over all folders in the out directory
221 |     for folder in subfolders:
222 |         tmp_fname = folder.split('/')
223 |         fname = tmp_fname[-1]
224 |         cl_id = fname
225 |         unique_cl_ids.add(cl_id)
226 |     # enter the folder containing all output for each cluster
227 |     for cl_id in unique_cl_ids:
228 |         #print(cl_id)
229 |         all_infos_dict = {}
230 |         batch_reads = {}
231 |         batch_mappings = {}
232 |         all_reads_dict = {}
233 |         cl_dir = os.path.join(outdir, cl_id)
234 |         # iterate over all batchfiles that were generated into the cluster's folder
235 |         for batchfile in os.listdir(cl_dir):
236 |             if batchfile.endswith("_batch"):
237 |                 tmp_bname = batchfile.split('/')
238 |                 tmp_bname2 = tmp_bname[-1].split('_')
239 |                 batch_id = int(tmp_bname2[0])
240 |                 #print(batch_id)
241 |                 batchfilename = str(batch_id) + "_batch"
242 |                 batch_file = open(os.path.join(cl_dir, batchfilename), 'rb')
243 |                 all_reads_dict[batch_id] = pickle.load(batch_file)
244 |                 for key in all_reads_dict.keys():
245 |                     all_infos_dict[key] = {}
246 |                 spoa_name = "spoa" + str(batch_id)
247 |                 #print(os.path.join(cl_dir, spoa_name))
248 |                 spoa_file = open(os.path.join(cl_dir, spoa_name), 'rb')
249 |                 batch_reads[batch_id] = pickle.load(spoa_file)
250 |                 map_name = "mapping" + str(batch_id)
251 |                 map_file = open(os.path.join(cl_dir, map_name), 'rb')
252 |                 batch_mappings[batch_id] = pickle.load(map_file)
253 |         # collect the information so that we have one data structure containing all infos
254 |         for b_id, value in batch_reads.items():
255 |             all_infos_batch = {}
256 |             for cons_id, seq in value.items():
257 |                 bm_this_batch = batch_mappings[b_id]
258 |                 reads = bm_this_batch[cons_id]
259 |                 read_mapping = Read(seq, reads, False)
260 |                 all_infos_batch[cons_id] = read_mapping
261 |             all_infos_dict[b_id] = all_infos_batch
262 | 
263 |         nr_reads = 0
264 |         for b_id, b_infos in all_infos_dict.items():
265 |             for c_id, c_infos in b_infos.items():
266 |                 if not c_infos.merged:
267 |                     nr_reads += len(c_infos.reads)
268 |         # perform the merging step during which all consensuses are compared and if possible merged
269 |         actual_merging_process(all_infos_dict, delta, delta_len,
270 |                                delta_iso_len_3, delta_iso_len_5, max_seqs_to_spoa, all_reads_dict, outdir)
271 |         nr_reads = 0
272 |         for b_id, b_infos in all_infos_dict.items():
273 |             for c_id, c_infos in b_infos.items():
274 |                 if not c_infos.merged:
275 |                     nr_reads += len(c_infos.reads)
276 |             write_final_output(all_infos_dict, outdir, iso_abundance, cl_dir, cl_id, write_fastq, write_low_abundance)
277 |         #shutil.rmtree(os.path.join(outdir,cl_id))
278 | 
279 | DEBUG = False
280 | 
281 | 
282 | def main():
283 |     # outfolder= "/home/alexanderpetri/isONform_analysis/Paraout_500_cl0"
284 |     # outfolder = "/home/alexanderpetri/isONform_analysis/Para_out_500batchadd"
285 |     # outfolder="/home/alexanderpetri/isONform_analysis/Batch_parallel_testing"
286 |     # outfolder = "/home/alexanderpetri/isONform_analysis/Para_out_500batch"
287 |     outfolder = "/home/alexanderpetri/Desktop/IsONform_test_results/SIRV100k/552997d_t2"
288 |     delta = 0.15
289 |     delta_len = 5
290 |     merge_sub_isoforms_3 = True
291 |     merge_sub_isoforms_5 = True
292 |     delta_iso_len_3 = 30
293 |     delta_iso_len_5 = 50
294 |     max_seqs_to_spoa = 200
295 |     iso_abundance = 4
296 |     join_back_via_batch_merging(outfolder, delta, delta_len, delta_iso_len_3, delta_iso_len_5,
297 |                                 max_seqs_to_spoa, iso_abundance)
298 | 
299 |     Parallelization_side_functions.generate_full_output(outfolder)
300 |     # merge_batches(max_batchid, tmp_work_dir, outfolder, all_reads,merge_sub_isoforms_3,merge_sub_isoforms_5, delta, delta_len,max_seqs_to_spoa, delta_iso_len_3, delta_iso_len_5,iso_abundance)
301 |     print("removing temporary workdir")
302 |     # shutil.rmtree(work_dir)
303 | 
304 | 
305 | if __name__ == "__main__":
306 |     main()
307 | 


--------------------------------------------------------------------------------
/isONform_parallel:
--------------------------------------------------------------------------------
  1 | #! /usr/bin/env python
  2 | """
  3 | This file is a wrapper file to run isONform in parallel (parallelization over batches).
  4 | The script was taken from isoncorrect (https://github.com/ksahlin/isoncorrect/blob/master/run_isoncorrect)
  5 | by Kristoffer Sahlin and changed by Alexander Petri to be usable with the isONform code base.
  6 | 
  7 | """
  8 | 
  9 | 
 10 | from __future__ import print_function
 11 | import argparse
 12 | import tempfile
 13 | from ast import Param
 14 | from time import time
 15 | from pathlib import Path
 16 | import signal
 17 | from multiprocessing import Pool
 18 | import multiprocessing as mp
 19 | 
 20 | import subprocess
 21 | import sys
 22 | import os
 23 | from sys import stdout
 24 | import shutil
 25 | import errno
 26 | 
 27 | from modules import batch_merging_parallel
 28 | from modules import help_functions
 29 | from modules import Parallelization_side_functions
 30 | 
 31 | def restructure_isoncorrect_output(directory):
 32 |     file_list= list(filter(lambda f: os.path.isfile(os.path.join(directory, f)), os.listdir(directory) ))
 33 |     #print("file_list:", file_list)
 34 |     if len(file_list) == 0:
 35 |         print("isONcorrect structure")
 36 |         #iterate over all subdirectories in directory
 37 |         for subdirectory in os.listdir(directory):
 38 |             subdirectory_path = os.path.join(directory, subdirectory)
 39 |             # Check if it's a directory
 40 |             if os.path.isdir(subdirectory_path):
 41 |                 # Define the source file path
 42 |                 source_file = os.path.join(subdirectory_path, 'corrected_reads.fastq')
 43 |                 if source_file:
 44 |                     # Define the target file path in the main folder with new name
 45 |                     target_file = os.path.join(directory, f"{subdirectory}.fastq")
 46 |                     # Move and rename if the file exists
 47 |                     if os.path.exists(source_file):
 48 |                         shutil.move(source_file, target_file)
 49 |                         print(f"Moved and renamed {source_file} to {target_file}")
 50 |                     else:
 51 |                         print(f"File {source_file} does not exist.")
 52 |         # The following line removes all folders in the input folder
 53 |         Parallelization_side_functions.remove_folders(directory)
 54 |     else:
 55 |         print("isONclust structure")
 56 | 
 57 | def wccount(filename):
 58 |     out = subprocess.Popen(['wc', '-l', filename],
 59 |                            stdout=subprocess.PIPE,
 60 |                            stderr=subprocess.STDOUT
 61 |                            ).communicate()[0]
 62 |     # print(int(out.split()[0]))
 63 |     return int(out.split()[0])
 64 | 
 65 | 
 66 | def isONform(data):
 67 |     isONform_location, read_fastq_file, outfolder, batch_id, isONform_algorithm_params,cl_id = data[0], data[1], data[
 68 |         2], data[3], data[4], data[5]
 69 |     help_functions.mkdir_p(outfolder)
 70 |     #print("OUT",outfolder)
 71 |     #print("Algoparams",isONform_algorithm_params)
 72 |     isONform_exec = os.path.join(isONform_location, "main")
 73 |     isONform_error_file = os.path.join(outfolder, "stderr.txt")
 74 |     with open(isONform_error_file, "w") as error_file:
 75 |         print('Running isONform batch_id:{0}.{1}...'.format(cl_id,batch_id), end=' ')
 76 |         stdout.flush()
 77 |         isONform_out_file = open(os.path.join(outfolder, "stdout{0}_{1}.txt".format(cl_id,batch_id)), "w")
 78 |         subprocess.check_call(
 79 |                 ["python", isONform_exec, "--fastq", read_fastq_file, "--outfolder", outfolder,
 80 |                  "--exact_instance_limit", str(isONform_algorithm_params["exact_instance_limit"]),
 81 |                  #"--max_seqs", str(isONform_algorithm_params["max_seqs"]),
 82 |                  "--k", str(isONform_algorithm_params["k"]), "--w", str(isONform_algorithm_params["w"]),
 83 |                  "--xmin", str(isONform_algorithm_params["xmin"]), "--xmax",
 84 |                  str(isONform_algorithm_params["xmax"]),"--delta_len", str(isONform_algorithm_params["delta_len"]),
 85 |                  "--exact", "--parallel", "True",  "--delta_iso_len_3", str(isONform_algorithm_params["delta_iso_len_3"]), "--delta_iso_len_5", str(isONform_algorithm_params["delta_iso_len_5"])#, "--slow"
 86 |                  #"--T", str(isONform_algorithm_params["T"])
 87 |                  ], stderr=error_file, stdout=isONform_out_file)
 88 | 
 89 |         print('Done with batch_id:{0}.{1}'.format(cl_id,batch_id))
 90 |         stdout.flush()
 91 |     error_file.close()
 92 |     isONform_out_file.close()
 93 |     return batch_id
 94 | 
 95 | #splits files containing more than max_seqs reads into smaller files, that can be parallelized upon
 96 | def splitfile(indir, tmp_outdir, fname, chunksize,cl_id,ext):
 97 |     # from https://stackoverflow.com/a/27641636/2060202
 98 |     # fpath, fname = os.path.split(infilepath)
 99 |     #cl_id, ext = fname.rsplit('.',1)
100 |     infilepath = os.path.join(indir, fname)
101 |     #infilepath=indir
102 |     # print(fpath, cl_id, ext)
103 |     #print("now at splitfile")
104 |     #print(indir, tmp_outdir, cl_id, ext)
105 | 
106 |     i = 0
107 |     written = False
108 |     with open(infilepath) as infile:
109 |         while True:
110 |             outfilepath = os.path.join(tmp_outdir, '{0}_{1}.{2}'.format(cl_id, i, ext) ) #"{}_{}.{}".format(foutpath, fname, i, ext)
111 |             #print(outfilepath)
112 |             with open(outfilepath, 'w') as outfile:
113 |                 for line in (infile.readline() for _ in range(chunksize)):
114 |                     outfile.write(line)
115 |                 written = bool(line)
116 |             # print(os.stat(outfilepath).st_size == 0)
117 |             if os.stat(outfilepath).st_size == 0: # Corner case: Original file is even multiple of max_seqs, hence the last file becomes empty. Remove this
118 |                 os.remove(outfilepath)
119 |             if not written:
120 |                 break
121 |             i += 1
122 | 
123 | 
124 | def symlink_force(target, link_name):
125 |     #print("Symlink",link_name)
126 |     try:
127 |         os.symlink(target, link_name)
128 |     except OSError as e:
129 |         if e.errno == errno.EEXIST:
130 |             if not os.path.exists(os.readlink(link_name)):
131 |                 print('path %s is a broken symlink' % link_name)
132 |                 os.remove(link_name)
133 |                 symlink_force(target,link_name)
134 |         else:
135 |             raise e
136 | 
137 | #splits clusters up so that we get smaller batches
138 | def split_cluster_in_batches_corrected(indir, outdir, tmp_work_dir, max_seqs):
139 |     # create a modified indir
140 |     tmp_work_dir = os.path.join(tmp_work_dir, 'split_in_batches')
141 |     # print(indir)
142 |     help_functions.mkdir_p(tmp_work_dir)
143 |     pat = Path(indir)
144 |     #collect all fastq files located in this directory or any subdirectories
145 |     file_list = list(pat.rglob('*.fastq'))
146 |     #print("FLIST",file_list)
147 |     #iterate over the fastq_files
148 |     for filepath in file_list:
149 |         #print("FPATH",filepath)
150 |         old_fastq_file = str(filepath.resolve())
151 |         path_split = old_fastq_file.split("/")
152 |         folder = path_split[-2]
153 |         #print(folder)
154 |         fastq_file = path_split[-1]
155 |         #we do not want to look at the analysis fastq file
156 |         if not folder == "Analysis":
157 |             cl_id=path_split[-2]
158 |             #print("CLID",cl_id)
159 | 
160 |             #if we have more lines than max_seqs
161 |             new_indir=os.path.join(indir,folder)
162 |             #print(new_indir)
163 | 
164 |             num_lines = sum(1 for line in open(os.path.join(new_indir, fastq_file)))
165 |             #print("Number Lines", fastq_file, num_lines)
166 | 
167 |             # determine whether the file is larger than max_seqs
168 |             larger_than_max_seqs = num_lines > 4 * max_seqs
169 |             if larger_than_max_seqs:
170 |                 #print("Splitting",filepath)
171 |                 ext = fastq_file.rsplit('.', 1)[1]
172 |                 splitfile(new_indir, tmp_work_dir, fastq_file, 4 * max_seqs,cl_id,ext)  # is fastq file
173 |             else:
174 |                 ext = fastq_file.rsplit('.', 1)[1]
175 |                 #print(fastq_file, "symlinking instead")
176 |                 symlink_force(filepath, os.path.join(tmp_work_dir, '{0}_{1}.{2}'.format(cl_id, 0, ext)))
177 |     return tmp_work_dir
178 | 
179 | def split_cluster_in_batches_clust(indir, outdir, tmp_work_dir, max_seqs):
180 |     # create a modified indir
181 |     tmp_work_dir = os.path.join(tmp_work_dir, 'split_in_batches')
182 |     # print(indir)
183 |     #os.mkdir(tmp_work_dir)
184 |     help_functions.mkdir_p(tmp_work_dir)
185 |     #print(tmp_work_dir)
186 |     #print("clust")
187 | 
188 |     pat = Path(indir)
189 |     file_list = list(pat.rglob('*.fastq'))
190 |     # add split files to this indir
191 |     for file_ in file_list:
192 |     #for file_ in sorted(os.listdir(indir), key=lambda x: int(x.split('.')[0])):
193 |         #fastq_path = os.fsdecode(file_)
194 |         old_fastq_file = str(file_.resolve())
195 |         fastq_file = old_fastq_file.split("/")[-1]
196 |         #print("FASTQ",fastq_file)
197 | 
198 |         num_lines = sum(1 for line in open(os.path.join(indir, fastq_file)))
199 | 
200 |         # determine whether the file is larger than max_seqs
201 |         larger_than_max_seqs = num_lines > 4 * max_seqs
202 |         if larger_than_max_seqs:
203 |             cl_id, ext = fastq_file.rsplit('.', 1)
204 |             splitfile(indir, tmp_work_dir, fastq_file, 4 * max_seqs, cl_id, ext) # is fastq file
205 |         else:
206 |             cl_id, ext = fastq_file.rsplit('.', 1)
207 |             print("SYMLINK",os.path.join(tmp_work_dir, '{0}_{1}.{2}'.format(cl_id, 0, ext)))
208 |             symlink_force(file_, os.path.join(tmp_work_dir, '{0}_{1}.{2}'.format(cl_id, 0, ext)))
209 |     return tmp_work_dir
210 | 
211 | 
212 | #PYTHONHASHSEED = 0
213 | def main(args):
214 |     globstart = time()
215 |     directory = args.fastq_folder
216 |     #Please note that this alters the output of isONcorrect i.e. the structure of corrected in the isONpipeline
217 |     restructure_isoncorrect_output(directory)
218 | 
219 | 
220 |     #print("MERGE?", args.merge_sub_isoforms_3, args.merge_sub_isoforms_5)
221 | 
222 |       # os.fsencode(args.fastq_folder)
223 |     write_low_abundance = False
224 |     #print(directory)
225 |     #print("ARGS",args)
226 |     isONform_location = os.path.dirname(os.path.realpath(__file__))
227 |     if args.split_wrt_batches:
228 |         if args.tmpdir:
229 |             tmp_work_dir = args.tmpdir
230 |             #curr_work_dir = os.getcwd()
231 |             #os.chdir(tmp_work_dir)
232 |             #try:
233 |                 # Create the directory
234 |                 #os.makedirs(tmp_work_dir)
235 |             #    print("Directory created successfully.")
236 |             #except FileExistsError:
237 |             #    print("Directory already exists.")
238 |             Parallelization_side_functions.mkdir_p(tmp_work_dir)
239 |             #os.chdir(curr_work_dir)
240 |         else:
241 |             tmp_work_dir = tempfile.mkdtemp()
242 |         print("Temporary workdirectory:", tmp_work_dir)
243 |         split_tmp_directory = split_cluster_in_batches_clust(directory, args.outfolder, tmp_work_dir,
244 |                                                                      args.max_seqs)
245 |         split_directory = os.fsencode(split_tmp_directory)
246 |     else:
247 | 
248 |         split_directory = os.fsencode(directory)
249 |         print("SplitDIR", split_directory)
250 |     instances = []
251 |     for file_ in os.listdir(split_directory):
252 |             #print(file_)
253 |             read_fastq_file = os.fsdecode(file_)
254 |             if read_fastq_file.endswith(".fastq"):
255 |                 #print("True")
256 |                 tmp_id = read_fastq_file.split(".")[0]
257 |                 snd_tmp_id = tmp_id.split("_")
258 |                 cl_id = snd_tmp_id[0]
259 |                 batch_id = snd_tmp_id[1] if len(snd_tmp_id) > 1 else 0
260 |                 outfolder = os.path.join(args.outfolder, cl_id)
261 |                 #print(batch_id,cl_id)
262 |                 #print(outfolder)
263 |                 fastq_file_path = os.path.join(os.fsdecode(split_directory), read_fastq_file)
264 |                 #print(fastq_file_path)
265 |                 compute = True
266 |                 if args.keep_old:
267 |                     candidate_corrected_file = os.path.join(outfolder, "isoforms.fastq")
268 |                     if os.path.isfile(candidate_corrected_file):
269 |                         if wccount(candidate_corrected_file) == wccount(fastq_file_path):
270 |                             #print("already computed cluster and complete file", batch_id)
271 |                             compute = False
272 | 
273 |                 if compute:
274 |                     #print("computing")
275 |                     isONform_algorithm_params = {"set_w_dynamically": args.set_w_dynamically,
276 |                                                     "exact_instance_limit": args.exact_instance_limit,
277 |                                                     "delta_len": args.delta_len,"--exact": True,
278 |                                                     "k": args.k, "w": args.w, "xmin": args.xmin, "xmax": args.xmax,
279 |                                                      "max_seqs": args.max_seqs, "parallel": True, "--slow": True, "delta_iso_len_3": args.delta_iso_len_3,
280 |                                                  "delta_iso_len_5": args.delta_iso_len_5}
281 |                     instances.append(
282 |                         (isONform_location, fastq_file_path, outfolder, batch_id, isONform_algorithm_params, cl_id))
283 |             else:
284 |                 continue
285 |     instances.sort(key=lambda x: (int(x[5]), int(x[3])))  # sorting on cluster_id and then batch_id numerically
286 |     print("Printing instances")
287 |     for t in instances:
288 |         print(t)
289 |     original_sigint_handler = signal.signal(signal.SIGINT, signal.SIG_IGN)
290 |     signal.signal(signal.SIGINT, original_sigint_handler)
291 |     try:
292 |         mp.set_start_method('spawn')
293 |     except RuntimeError:
294 |         pass
295 |     #mp.set_start_method('spawn')
296 |     print(mp.get_context())
297 |     print("Environment set:", mp.get_context())
298 |     print("Using {0} cores.".format(args.nr_cores))
299 |     start_multi = time()
300 |     pool = Pool(processes=int(args.nr_cores))
301 |     try:
302 |         start = time()
303 |         for x in pool.imap_unordered(isONform, instances):
304 |             print("{} (Time elapsed: {}s)".format(x, int(time() - start)))
305 |     except KeyboardInterrupt:
306 |         print("Caught KeyboardInterrupt, terminating workers")
307 |         pool.terminate()
308 |         sys.exit()
309 |     else:
310 |         pool.close()
311 |     pool.join()
312 |     print("Time elapsed multiprocessing:", time() - start_multi)
313 | 
314 |     print("Merging...")
315 |     file_handling = time()
316 |     if args.write_fastq:
317 |         write_fastq = True
318 |     else:
319 |         write_fastq = False
320 |     batch_merging_parallel.join_back_via_batch_merging(args.outfolder, args.delta, args.delta_len, args.delta_iso_len_3, args.delta_iso_len_5, args.max_seqs_to_spoa, args.iso_abundance, write_fastq, write_low_abundance)
321 |     Parallelization_side_functions.generate_full_output(args.outfolder, write_fastq, write_low_abundance)
322 |     Parallelization_side_functions.remove_folders(args.outfolder)
323 |     if args.split_wrt_batches:
324 |         print("Removed the split directory")
325 |         shutil.rmtree(split_directory)
326 |     print("Joined back batched files in:", time() - file_handling)
327 |     print("Finished full algo after :", time() - globstart)
328 |     return
329 | 
330 | 
331 | if __name__ == '__main__':
332 |     parser = argparse.ArgumentParser(description="De novo reconstruction of long-read transcriptome reads",
333 |                                      formatter_class=argparse.ArgumentDefaultsHelpFormatter)
334 |     parser.add_argument('--version', action='version', version='%(prog)s 0.3.9')
335 |     parser.add_argument('--fastq_folder', type=str, default=False,
336 |                         help='Path to input fastq folder with reads in clusters')
337 |     parser.add_argument('--t', dest="nr_cores", type=int, default=8, help='Number of cores allocated for clustering')
338 |     parser.add_argument('--k', type=int, default=20, help='Kmer size')
339 |     parser.add_argument('--w', type=int, default=31, help='Window size')
340 |     parser.add_argument('--xmin', type=int, default=18, help='Lower interval length')
341 |     parser.add_argument('--xmax', type=int, default=80, help='Upper interval length')
342 |     parser.add_argument('--exact_instance_limit', type=int, default=50,
343 |                         help='Do exact correction for clusters under this threshold')
344 |     parser.add_argument('--keep_old', action="store_true",
345 |                         help='Do not recompute previous results if corrected_reads.fq is found and has the smae number of reads as input file (i.e., is complete).')
346 |     parser.add_argument('--set_w_dynamically', action="store_true",
347 |                         help='Set w = k + max(2*k, floor(cluster_size/1000)).')
348 |     parser.add_argument('--max_seqs', type=int, default=1000,
349 |                         help='Maximum number of seqs to correct at a time (in case of large clusters).')
350 |     parser.add_argument('--split_wrt_batches', action="store_true",
351 |                         help='Process reads per batch (of max_seqs sequences) instead of per cluster. Significantly decrease runtime when few very large clusters are less than the number of cores used.')
352 |     #parser.add_argument('--clustered', action="store_true",
353 |     #                    help='Indicates whether we use the output of isONclust (i.e. we have uncorrected data)')
354 |     parser.add_argument('--outfolder', type=str, default=None, help='Outfolder with all corrected reads.')
355 |     parser.add_argument('--delta_len', type=int, default=5,
356 |                         help='Maximum length difference between two reads intervals for which they would still be merged')
357 |     parser.add_argument('--delta',type=float,default=0.1, help='diversity rate used to compare sequences')
358 |     parser.add_argument('--max_seqs_to_spoa', type=int, default=200, help='Maximum number of seqs to spoa')
359 |     parser.add_argument('--verbose', action="store_true", help='Print various developer stats.')
360 |     parser.add_argument('--iso_abundance', type=int, default=5,
361 |                         help='Cutoff parameter: abundance of reads that have to support an isoform to show in results')
362 |     parser.add_argument('--delta_iso_len_3', type=int, default=30,
363 |                         help='Cutoff parameter: maximum length difference at 3prime end, for which subisoforms are still merged into longer isoforms')
364 |     parser.add_argument('--delta_iso_len_5', type=int, default=50,
365 |                         help='Cutoff parameter: maximum length difference at 5prime end, for which subisoforms are still merged into longer isoforms')
366 |     parser.add_argument('--tmpdir', type=str,default=None, help='OPTIONAL PARAMETER: Absolute path to custom folder in which to store temporary files. If tmpdir is not specified, isONform will attempt to write the temporary files into the tmp folder on your system. It is advised to only use this parameter if the symlinking does not work on your system.')
367 |     parser.add_argument('--write_fastq', action="store_true", help=' Indicates that we want to ouptut the final output (transcriptome) as fastq file (New standard: fasta)')
368 | 
369 |     args = parser.parse_args()
370 |     print(len(sys.argv))
371 |     if len(sys.argv) == 1:
372 |         parser.print_help()
373 |         sys.exit()
374 | 
375 | 
376 |     if args.outfolder and not os.path.exists(args.outfolder):
377 |         os.makedirs(args.outfolder)
378 | 
379 |     main(args)
380 | 


--------------------------------------------------------------------------------
/modules/IsoformGeneration.py:
--------------------------------------------------------------------------------
  1 | import _pickle as pickle
  2 | import itertools
  3 | import os
  4 | from sys import stdout
  5 | import subprocess
  6 | 
  7 | from modules import consensus
  8 | 
  9 | def write_consensus_file(batch_id, outfolder, new_consensuses):
 10 |     #print("newconsensus")
 11 |     #for con,seq in new_consensuses.items():
 12 |     #    print("{0}:{1}".format(con,seq))
 13 |     consensus_name = "spoa" + str(batch_id)
 14 |     pickle_spoa_file = open(os.path.join(outfolder, consensus_name), 'wb')
 15 |     pickle.dump(new_consensuses, pickle_spoa_file)
 16 |     pickle_spoa_file.close()
 17 | 
 18 | 
 19 | def write_mapping_file(batch_id, mapping, outfolder):
 20 |     mapping_name = "mapping" + str(batch_id)
 21 |     pickle_map_file = open(os.path.join(outfolder, mapping_name), 'wb')
 22 |     pickle.dump(mapping, pickle_map_file)
 23 |     pickle_map_file.close()
 24 | 
 25 | 
 26 | 
 27 | def prepare_consensuses(new_consensuses,equal_reads_keys, final_consensuses):
 28 |     for key in equal_reads_keys:
 29 |         final_consensuses[key] = new_consensuses[key]
 30 | 
 31 | 
 32 | def write_isoforms_pickle(equal_reads, reads, outfolder, batch_id, new_consensuses):
 33 |     print("Generating the Isoforms-merged")
 34 |     mapping = {}
 35 |     final_consensuses = {}
 36 |     prepare_consensuses(new_consensuses, equal_reads.keys(), final_consensuses)
 37 |     write_consensus_file(batch_id, outfolder, final_consensuses)
 38 |     # we iterate over all items in curr_best_seqs
 39 |     for key, value in equal_reads.items():
 40 |         name = str(key)
 41 |         # We generate a list as value for mapping[key]
 42 |         if name not in mapping:
 43 |             mapping[key] = []
 44 |         # iterate over all reads in value and add them to mapping
 45 |         for i, q_id in enumerate(value):
 46 |             singleread = reads[q_id]
 47 |             mapping[key].append(singleread[0])
 48 |     write_mapping_file(batch_id, mapping, outfolder)
 49 | 
 50 | 
 51 | 
 52 | 
 53 | def is_Sublist(l, s):
 54 |     """Function that decides whether a list 's' is a sublist of list 'l'.
 55 |     taken from: https://www.w3resource.com/python-exercises/list/python-data-type-list-exercise-32.php
 56 |     INPUT:      l:          the list
 57 |                 s:          the potential sublist
 58 |     OUTPUT      sub_set:    boolean value indicating the result
 59 |     """
 60 |     sub_set = False
 61 |     if not s:
 62 |         sub_set = True
 63 |     elif s == l:
 64 |         sub_set = True
 65 |     elif len(s) > len(l):
 66 |         sub_set = False
 67 |     else:
 68 |         for i in range(len(l)):
 69 |             if l[i] == s[0]:
 70 |                 n = 1
 71 |                 while (n < len(s)) and (l[i + n] == s[n]):
 72 |                     n += 1
 73 | 
 74 |                 if n == len(s):
 75 |                     sub_set = True
 76 | 
 77 |     return sub_set
 78 | 
 79 | 
 80 | def merge_isoform_paths(isoforms):
 81 |     """
 82 |     Merges subisoforms into larger isoforms
 83 |     """
 84 |     merge_list = []
 85 |     for id1, vis_nodes_set1 in isoforms.items():
 86 |         for id2, vis_nodes_set2 in isoforms.items():
 87 |             if id1<id2:
 88 |                 if is_Sublist(vis_nodes_set2,vis_nodes_set1):
 89 |                     merge_tuple = (id1,id2)
 90 |                     merge_list.append(merge_tuple)
 91 |                 elif is_Sublist(vis_nodes_set1,vis_nodes_set2):
 92 |                     merge_tuple = (id2,id1)
 93 |                     merge_list.append(merge_tuple)
 94 |     for tup in merge_list:
 95 |         subiso = tup[0]
 96 |         largeiso = tup[1]
 97 |         isoforms[largeiso].extend(isoforms[subiso])
 98 |         del isoforms[subiso]
 99 | 
100 | 
101 | def compute_equal_reads(DG, support, equal_reads):
102 |     """Method to generate the final isoforms by iterating through the graph structure
103 |     INPUT:      DG          Directed Graph
104 |                 support       list of reads
105 |     OUPUT:      filename    file which contains all the final isoforms
106 |     """
107 |     #path_and_support will hold the infos concerning the found paths
108 |     node_support_left = set(support)
109 |     #indicates whether we want to merge a true subisoform into another isoform (ie. the isoform is a continous sublist of the longer one)
110 |     merge_sub_isos=True
111 |     #we iterate as long as still not all support was allocated to a path
112 |     while node_support_left:
113 |         #print("Node_support_left",node_support_left)
114 |         node = "s"
115 |         #current_node_support = node_support_left
116 |         read=node_support_left.pop()
117 |         current_node_support = node_support_left
118 |         current_node_support.add(read)
119 |         visited_nodes = []
120 |         #As long as we have not visited the bubble end node we continue walking through our graph
121 |         while node != "t":
122 |             visited_nodes.append(node)
123 |             out_edges = DG.out_edges(node)
124 |             next_found = False
125 |             for edge in out_edges:
126 |                 edge_supp = DG[edge[0]][edge[1]]['edge_supp']
127 |                 if read in edge_supp:
128 |                     node = edge[1]
129 |                     current_node_support = current_node_support.intersection(edge_supp)
130 |                     next_found = True
131 |                     break
132 |             if not next_found:
133 |                 break
134 |         if current_node_support:
135 |             id = list(current_node_support)[0]
136 |             equal_reads[id] = list(current_node_support)
137 |             node_support_left -= current_node_support
138 |     if merge_sub_isos:
139 |         merge_isoform_paths(equal_reads)
140 |     return equal_reads
141 | 
142 | 
143 | def run_spoa(reads, spoa_out_file):
144 |     """calls spoa and returns the consensus sequence for the given reads"""
145 |     with open(spoa_out_file, "w") as output_file:
146 |         stdout.flush()
147 |         null = open("/dev/null", "w")
148 |         subprocess.check_call(["spoa", reads, "-l", "0", "-r", "0", "-g", "-2"],
149 |                                   stdout=output_file, stderr=null)
150 |         stdout.flush()
151 |     l = open(spoa_out_file, "r").readlines()
152 |     output_file.close()
153 |     consensus = l[1].strip()
154 |     del l
155 |     null.close()
156 |     return consensus
157 | 
158 | 
159 | def generate_isoform_using_spoa(curr_best_seqs,reads, work_dir,outfolder,batch_id,iso_abundance, max_seqs_to_spoa=200):
160 |     """retreives the sequences which are to be aligned using spoa and writes the consensus into a file
161 | 
162 |         curr_best_seqs is an array with q_id, pos1, pos2
163 |         the current read is on index 0 in curr_best_seqs array
164 |     """
165 |     print("Generating the Isoforms")
166 |     mapping = {}
167 |     consensus_name="spoa"+str(batch_id)+"merged.fasta"
168 |     consensus_file = open(os.path.join(outfolder, consensus_name), 'w')
169 |     for key,value in curr_best_seqs.items():
170 |         name = 'consensus' + str(value[0])
171 |         mapping[name] = []
172 |         reads_path = open(os.path.join(work_dir, "reads_tmp.fa"), "w")
173 |         if len(value)>=iso_abundance:
174 |             if len(value) == 1:
175 |                 rid = key
176 |                 singleread = reads[rid]
177 |                 seq = singleread[1]
178 |                 mapping[name].append(singleread[0])
179 |                 consensus_file.write(">{0}\n{1}\n".format(name, seq))
180 |                 reads_path.close()
181 |             else:
182 |                 for i, q_id in enumerate(value):
183 |                     singleread = reads[q_id]
184 |                     seq = singleread[1]
185 |                     mapping[name].append(singleread[0])
186 |                     if i<max_seqs_to_spoa:
187 |                         reads_path.write(">{0}\n{1}\n".format(singleread[0], seq))
188 | 
189 |                 reads_path.close()
190 |                 spoa_ref = run_spoa(reads_path.name, os.path.join(work_dir, "spoa_tmp.fa"))
191 |                 consensus_file.write(">{0}\n{1}\n".format(name, spoa_ref))
192 | 
193 |     mapping_name="mapping"+str(batch_id)+".txt"
194 |     mappingfile = open(os.path.join(outfolder, mapping_name), "w")
195 |     # we now write the information into the mappingfile
196 |     for id, readlist in mapping.items():
197 |         if len(readlist)>=iso_abundance:
198 |             mappingfile.write("{0}\n{1}\n".format(id, readlist))
199 | 
200 |     #close the files we wrote into
201 |     mappingfile.close()
202 |     consensus_file.close()
203 | 
204 | 
205 | 
206 | 
207 | def search_last_entries(entry_since_sign_match,delta_len_3):
208 |     dist_to_last_match = 0
209 |     deletion_len = 0
210 |     insertion_len = 0
211 |     for entry in entry_since_sign_match:
212 |         cig_len = entry[0]
213 |         cig_type = entry[1]
214 |         dist_to_last_match += cig_len
215 |         if cig_type == "D":
216 |             deletion_len += cig_len
217 |         if cig_type == "I":
218 |             insertion_len += cig_len
219 |     #if DEBUG:
220 |         #print(dist_to_last_match)
221 |         #print(deletion_len)
222 |     if dist_to_last_match - deletion_len < delta_len_3:
223 |         return True
224 |     else:
225 |         return False
226 | 
227 | 
228 | def search_first_entries(before_fsm,first_sign_match,delta_len_5):
229 |     deletion_len = 0
230 |     for entry in before_fsm:
231 |         cig_len = entry[0]
232 |         cig_type = entry[1]
233 |         if cig_type == "D":
234 |             deletion_len += cig_len
235 |     if DEBUG:
236 |         print(first_sign_match)
237 |         print(deletion_len)
238 |         print(delta_len_5)
239 |     if first_sign_match - deletion_len < delta_len_5:
240 |         if DEBUG:
241 |             print("res", first_sign_match - deletion_len < delta_len_5)
242 |         return True
243 |     else:
244 |         if DEBUG:
245 |             print("res", first_sign_match - deletion_len < delta_len_5)
246 |         return False
247 | 
248 | 
249 | 
250 | #parses the parasail alignment output to figure out whether to merge
251 | def parse_cigar_diversity_isoform_level_new(cigar_tuples, delta,delta_len,delta_iso_len_3,delta_iso_len_5,overall_len,first_match,last_match):
252 |     miss_match_length = 0
253 |     alignment_len = 0
254 |     after_last_matches=0
255 |     after_last_nomatch=0
256 |     before_first_matches=0
257 |     before_first_nomatch=0
258 |     for i, elem in enumerate(cigar_tuples):
259 |         #thisstartpos: the starting position of where the cigar tuple starts. (Previous end position)
260 |         this_start_pos=alignment_len
261 |         cig_len = elem[0]
262 |         cig_type = elem[1]
263 |         alignment_len += cig_len
264 |         #we found a mismatch
265 |         if (cig_type != '=') and (cig_type != 'M'):
266 |             #if the missmatch is located before the first significant match (upstream)
267 |             if this_start_pos < first_match:
268 |                 if not cig_type =='D':
269 |                     before_first_nomatch+=cig_len
270 | 
271 |             elif this_start_pos>=last_match or this_start_pos+cig_len>=last_match:
272 |                 if not cig_type =='D':
273 |                     after_last_nomatch+=cig_len
274 |             #the mismatch is located between the first and last significant matches
275 |             else:
276 |                 #if the mismatch is longer than delta_len: Structural difference -> not mergeable
277 |                 if cig_len > delta_len:
278 |                     return False
279 |                 #we still need this mismatch_length to be added to miss_match_length to document the number of mismatches between fsm and lsm
280 |                 else:
281 |                     # we want to add up all missmatches to compare to sequence length
282 |                     miss_match_length += delta_len
283 |         #we have a match
284 |         else:
285 |             if this_start_pos < first_match:
286 |                 before_first_matches += cig_len
287 |             elif this_start_pos >= last_match:
288 |                 after_last_matches+=cig_len
289 | 
290 |     #we know we have a match, but is it significant?
291 |     mergeable_start=before_first_matches+before_first_nomatch < delta_iso_len_5
292 |     #analyse the last entries of our cigar tuples to figure out what has happened after the lsm
293 |     mergeable_end=after_last_matches+after_last_nomatch< delta_iso_len_3
294 |     #the shorter sequence still went on longer than significant_match_len->not mergeable
295 |     if not mergeable_end or not mergeable_start:
296 |         if DEBUG:
297 |             print("NotMergeable start",mergeable_start," end:",mergeable_end)
298 |             print(cigar_tuples)
299 |         return False
300 |     #We calculate the diversity of our alignment
301 |     similar_seq=last_match-first_match
302 |     #just to make sure that we only merge reads that have at least 100 nt similar
303 |     if similar_seq<100:
304 |         #print("smaller sim_seq")
305 |         return False
306 |     diversity = (miss_match_length/ similar_seq)
307 |     if overall_len<200:
308 |         delta=2*delta
309 |     max_bp_diff = max(delta * similar_seq, delta_len)
310 |     mod_div_rate = max_bp_diff / similar_seq
311 |     #print("DIV",diversity,", mod_div_rate",mod_div_rate)
312 |     #we additionally make sure that the two consensuses are not too diverse
313 |     diversity_bool=diversity <= mod_div_rate
314 |     #if any of the three parameters we look at tells us not to merge we do not merge
315 |     if diversity_bool:  # delta_perc:
316 |         #print("Div:",diversity_bool,"3' ",three_prime," 5' ",five_prime)
317 |         return True
318 |     else:
319 |         #print("no pop due to diversity")
320 |         return False
321 | 
322 | 
323 | def parse_cigar_diversity_isoform_level(cigar_tuples, delta,delta_len,delta_iso_len_3,delta_iso_len_5,overall_len):
324 |     miss_match_length = 0
325 |     alignment_len = 0
326 |     for i, elem in enumerate(cigar_tuples):
327 |         this_start_pos=alignment_len
328 |         cig_len = elem[0]
329 |         cig_type = elem[1]
330 |         alignment_len += cig_len
331 |         if (cig_type != '=') and (cig_type != 'M'):
332 |             # we want to add up all missmatches to compare to sequence length
333 |             miss_match_length += cig_len
334 |             if cig_len <= delta_len:
335 |                 continue
336 |             # we also want to make sure that we do not have too large internal sequence differences
337 | 
338 |             #if the user wants us to merge the 5'end
339 |             #make sure that the startposition of the cigar tuple is in delta_iso_len
340 |             if this_start_pos < delta_iso_len_5:
341 |                 #we only want to merge if the cigar tuple does not span into the read by more than delta_len + delta_iso_len_5 base pairs
342 |                 if this_start_pos+cig_len<delta_iso_len_5+delta_len:
343 |                     continue
344 |             #we now want that the cigar string starts at most delta_iso_len+delta_len base pairs before the end of the read
345 |             if  this_start_pos > ((overall_len - delta_iso_len_3)-delta_len):
346 |                 continue
347 |             #the user wanted us to neither merge at 3' nor at 5', but we found a cigar tuple entry longer than delta_len indicating a missmatch
348 |             return False
349 | 
350 |     #We calculate the diversity of our alignment
351 |     diversity = (miss_match_length / alignment_len)
352 |     max_bp_diff = max(delta * alignment_len, delta_len)
353 |     mod_div_rate = max_bp_diff / alignment_len
354 |     #we additionally make sure that the two consensuses are not too diverse
355 |     diversity_bool=diversity <= mod_div_rate
356 |     #if any of the three parameters we look at tells us not to merge we do not merge
357 |     if diversity_bool:
358 |         return True
359 |     else:
360 |         return False
361 | 
362 | 
363 | def get_overall_alignment_len(cigar_tuples):
364 |     overall_len=0
365 |     for i, elem in enumerate(cigar_tuples):
366 |         overall_len += elem[0]
367 |     return overall_len
368 | 
369 | 
370 | def find_first_significant_match(s1_alignment, s2_alignment, windowsize, alignment_threshold):
371 |     match_vector = [1 if n1 == n2 else 0 for n1, n2 in zip(s1_alignment, s2_alignment)]
372 |     for i in range(0,len(match_vector)-windowsize+1):
373 |         our_equality = sum(match_vector[i:i+windowsize])/windowsize
374 |         #we only have a significant match if the equality of the covered area is larger than alignment_threshold
375 |         if our_equality > alignment_threshold:
376 |             return i
377 |     return -1
378 | 
379 | 
380 | def align_to_merge(consensus1,consensus2,delta,delta_len,delta_iso_len_3,delta_iso_len_5):
381 |     s1_alignment, s2_alignment, cigar_string, cigar_tuples, score = consensus.parasail_alignment(consensus1, consensus2,
382 |                                                                                        match_score=2,
383 |                                                                                        mismatch_penalty=-2,
384 |                                                                                        opening_penalty=12, gap_ext=1)
385 |     overall_len=get_overall_alignment_len(cigar_tuples)
386 |     windowsize=30
387 |     alignment_threshold=0.7
388 |     start_match=find_first_significant_match(s1_alignment,s2_alignment,windowsize,alignment_threshold)
389 |     end_match=find_first_significant_match(s1_alignment[::-1],s2_alignment[::-1],windowsize,alignment_threshold)
390 |     if start_match < 0 or end_match < 0:
391 |         return False
392 |     end_match_pos=overall_len-end_match
393 |     good_to_pop = parse_cigar_diversity_isoform_level_new(cigar_tuples, delta, delta_len, delta_iso_len_3, delta_iso_len_5, overall_len, start_match, end_match_pos)
394 |     if DEBUG:
395 |         print(cigar_tuples)
396 |         print(cigar_string)
397 |     return good_to_pop
398 | 
399 | 
400 | 
401 | def generate_new_full_consensus(id,id2,reads,curr_best_seqs,work_dir,max_seqs_to_spoa):
402 |     consensus_infos1 = curr_best_seqs[id]
403 |     consensus_infos2 = curr_best_seqs[id2]
404 |     reads_path = open(os.path.join(work_dir, "reads_tmp.fa"), "w")
405 |     all_consensus_infos=consensus_infos1+consensus_infos2
406 |     if len(all_consensus_infos) > max_seqs_to_spoa:
407 |         all_consensus_infos=all_consensus_infos[0:max_seqs_to_spoa]
408 |     for q_id in all_consensus_infos:
409 |         singleread = reads[q_id]
410 |         seq = singleread[1]
411 |         reads_path.write(">{0}\n{1}\n".format(singleread[0], seq))
412 |     reads_path.close()
413 |     spoa_ref = run_spoa(reads_path.name, os.path.join(work_dir, "spoa_tmp.fa"))
414 |     return spoa_ref
415 | 
416 | 
417 | def generate_all_consensuses(all_consensuses,alternative_consensuses,curr_best_seqs,reads,work_dir,max_seqs_to_spoa):
418 |     print("Generating")
419 |     for id,seqs in curr_best_seqs.items():
420 |         reads_path = open(os.path.join(work_dir, "reads_tmp.fa"), "w")
421 |         if len(seqs) == 1:
422 |             rid = id
423 |             singleread = reads[rid]
424 |             seq = singleread[1]
425 |             all_consensuses[id] = seq
426 |             consensus_tuple = (id, seq)
427 |             alternative_consensuses.append(consensus_tuple)
428 |             reads_path.close()
429 |         else:
430 |             for i, q_id in enumerate(seqs):
431 |                 singleread = reads[q_id]
432 |                 seq = singleread[1]
433 |                 if i < max_seqs_to_spoa:
434 |                     reads_path.write(">{0}\n{1}\n".format(singleread[0], seq))
435 |             reads_path.close()
436 |             spoa_ref = run_spoa(reads_path.name, os.path.join(work_dir, "spoa_tmp.fa"))
437 |             all_consensuses[id] = spoa_ref
438 |             consensus_tuple=(id,spoa_ref)
439 |             alternative_consensuses.append(consensus_tuple)
440 | 
441 | 
442 | def add_merged_reads(curr_best_seqs, id2,id1):
443 |     #we want to merge all sequences of id1 into id2
444 |     consensus_to_update=curr_best_seqs[id2]
445 |     consensus_updates=curr_best_seqs[id1]
446 |     #all the support of id1 and id2 are merged into new_consensus
447 |     new_consensus=consensus_to_update+consensus_updates
448 |     #add the new consensuses to curr_best_seqs for id2
449 |     curr_best_seqs[id2]=new_consensus
450 |     #we pop id1 from curr_best_Seqs as this id is not needed anymore
451 |     curr_best_seqs.pop(id1)
452 | 
453 | 
454 | 
455 | 
456 | """
457 | This method is used to find out how similar the consensuses are (by figuring out how much of their intervals are shared.
458 | INPUT: isoform_paths: List object of all nodes visited for each isoform
459 |         outfolder:         The folder in which we want to write our infos
460 |         batch_id:           The id of the current batch to be able to tell apart the infos of different batches
461 |     The method produces two files:  A similarity file giving the similarity values for each pair of consuensuses.
462 |                                     A path file giving the paths of all final isoforms
463 | """
464 | def merge_consensuses(curr_best_seqs,work_dir,delta,delta_len,reads,max_seqs_to_spoa,delta_iso_len_3,delta_iso_len_5):
465 |     new_consensuses={}
466 |     all_consensuses={}
467 |     alternative_consensuses=[]
468 |     #generate a consensus for each element in curr_best_seqs and save them in all_consensuses(dict: key-id,val-consensus) and alternative_consensuses
469 |     generate_all_consensuses(all_consensuses,alternative_consensuses,curr_best_seqs,reads,work_dir,max_seqs_to_spoa)
470 |     #sort alternative_consensuses by the length of the consensus sequences
471 |     alternative_consensuses.sort(key=lambda x: len(x[1]))
472 |     #iterate over alternative_consensuses (but always the shorter consensus)
473 |     for i, consensus in enumerate(alternative_consensuses[:len(alternative_consensuses)-1]):
474 |         id1=consensus[0]
475 |         if not id1 in new_consensuses:
476 |             seq1=consensus[1]
477 |         else:
478 |             seq1=new_consensuses[id1]
479 |         #iterate over the rest of the consensuses
480 |         for consensus2 in alternative_consensuses[i+1:]:
481 |             id2=consensus2[0]
482 |             if DEBUG:
483 |                 print(id1,id2)
484 | 
485 |             seq2=consensus2[1]
486 |             if len(seq2)<len(seq1):
487 |                 consensus1=seq2
488 |                 consensus2=seq1
489 |             else:
490 |                 consensus1 = seq1
491 |                 consensus2 = seq2
492 |             merge_consensuses_possible = align_to_merge(consensus1, consensus2, delta, delta_len, delta_iso_len_3, delta_iso_len_5)
493 |             if merge_consensuses_possible:
494 |                 first_consensus=[item for item in alternative_consensuses if item[0] == id2][0]
495 |                 if len(curr_best_seqs[id2]) > 50:
496 |                     add_merged_reads(curr_best_seqs, id2,id1)
497 |                     new_consensuses[id2]=first_consensus[1]
498 |                 else:
499 |                     new_consensuses[id2]=generate_new_full_consensus(id1,id2,reads,curr_best_seqs,work_dir,max_seqs_to_spoa)
500 |                     add_merged_reads(curr_best_seqs, id2, id1)
501 |                 break
502 |     for id,support in curr_best_seqs.items():
503 |         if not id in new_consensuses:
504 |             seq=all_consensuses[id]
505 |             new_consensuses[id]=seq
506 | 
507 |     return new_consensuses
508 | 
509 | 
510 | def generate_isoforms_new(equal_reads, reads, outfolder, batch_id,new_consensuses):
511 |     print("Generating the Isoforms-merged")
512 |     mapping = {}
513 |     consensus_name = "spoa" + str(batch_id) + "merged.fasta"
514 |     support_name = "support" + str(batch_id) + ".txt"
515 |     consensus_file = open(os.path.join(outfolder, consensus_name), 'w')
516 |     support_file = open(os.path.join(outfolder, support_name), 'w')
517 |     # we iterate over all items in curr_best_seqs
518 |     for key, value in equal_reads.items():
519 |         name = 'consensus' + str(key)
520 |         # We generate a list as value for mapping[key]
521 |         if name not in mapping:
522 |             mapping[name] = []
523 |         # iterate over all reads in value and add them to mapping
524 |         for i, q_id in enumerate(value):
525 |                     singleread = reads[q_id]
526 |                     mapping[name].append(singleread[0])
527 |                 # we do not have to calculate consensus as already done
528 |         sequence = new_consensuses[key]
529 |         consensus_file.write(">{0}\n{1}\n".format(name, sequence))
530 |         other_mapping_cter = 0
531 |         for key, value in mapping.items():
532 |             other_mapping_cter += len(value)
533 | 
534 |     # we have to treat the mapping file generation a bit different as we have to collect also the supporting reads of the subconsensuses
535 |     mapping_name = "mapping" + str(batch_id) + ".txt"
536 |     mappingfile = open(os.path.join(outfolder, mapping_name), "w")
537 |     for id, mapped_read_list in mapping.items():
538 |             mappingfile.write("{0}\n{1}\n".format(id, mapped_read_list))
539 |             support_file.write("{0}: {1}\n".format(id, len(mapped_read_list)))
540 |     consensus_file.close()
541 |     support_file.close()
542 |     mappingfile.close()
543 | 
544 | 
545 | 
546 | def generate_isoforms(DG,all_reads,reads,work_dir,outfolder,batch_id,delta,delta_len,delta_iso_len_3,delta_iso_len_5,max_seqs_to_spoa=200):
547 |     """
548 |     Wrapper method used for the isoform generation
549 |     """
550 |     equal_reads={}
551 |     compute_equal_reads(DG,reads,equal_reads)
552 |     if DEBUG == True:
553 |         print("EQUALS",equal_reads)
554 |     new_consensuses = merge_consensuses(equal_reads, work_dir, delta, delta_len,
555 |                                   all_reads, max_seqs_to_spoa,
556 |                                  delta_iso_len_3, delta_iso_len_5)
557 |     if DEBUG:
558 |             print("HELLO",new_consensuses)
559 |         #generate_isoforms_new(equal_reads, all_reads, outfolder, batch_id,new_consensuses)
560 |     write_isoforms_pickle(equal_reads, all_reads, outfolder, batch_id, new_consensuses)
561 | DEBUG = False
562 | 
563 | 
564 | def write_transcriptome_single(outfolder):
565 |     spoa_file = open(os.path.join(outfolder, "spoa0"), 'rb')
566 |     spoa_infos = pickle.load(spoa_file)
567 |     consensus_name="transcriptome.fastq"
568 |     consensus_file = open(os.path.join(outfolder, consensus_name), "w")
569 |     for id, sequence in spoa_infos.items():
570 |         inter_id = id.replace('\n', '')
571 |         new_id = int(inter_id.replace('>consensus', ''))
572 |         sequence = sequence.replace('\n', '')
573 |         consensus_file.write("@{0}\n{1}\n+\n{2}\n".format(new_id, sequence,
574 |                                                                   "+" * len(sequence)))
575 |     consensus_file.close()
576 | 
577 | 
578 | 
579 | def main():
580 |     outfolder="100kSIRV/20722_abundance2_par/22"
581 |     batch_id = 0
582 |     print("Hello World")
583 |     file = open(os.path.join(outfolder,'equal_reads_'+str(batch_id)+'.txt'), 'rb')
584 |     equal_reads = pickle.load(file)
585 |     file.close()
586 |     file = open(os.path.join(outfolder,'isoform_paths_'+str(batch_id)+'.txt'), 'rb')
587 |     isoform_paths = pickle.load(file)
588 |     file.close()
589 | 
590 |     file2 = open(os.path.join(outfolder, 'all_reads_' + str(batch_id)+".txt" ), 'rb')
591 |     #file2 = open(os.path.join(outfolder,'all_reads_'+str(batch_id)+'.txt'), 'rb')
592 |     all_reads = pickle.load(file2)
593 |     file2.close()
594 |     #work_dir = tempfile.mkdtemp()
595 |     outfolder = "out_local"
596 |     delta=0.10
597 |     delta_len=5
598 | 
599 |     merge_sub_isoforms_3=True
600 |     merge_sub_isoforms_5 = True
601 |     delta_iso_len_3=30
602 |     delta_iso_len_5 = 50
603 |     max_seqs_to_spoa=200
604 |     iso_abundance=1
605 | 
606 |     #if merge_sub_isoforms_5 or merge_sub_isoforms_3:
607 |     #    merged_dict,merged_consensuses,called_consensuses,consensus_map = calculate_isoform_similarity(equal_reads, work_dir, isoform_paths, outfolder, delta, delta_len, batch_id,
608 |     #                             merge_sub_isoforms_3, merge_sub_isoforms_5, all_reads, max_seqs_to_spoa,
609 |     #                             delta_iso_len_3, delta_iso_len_5)
610 |     #    generate_isoforms_new()
611 |     #    generate_isoform_using_spoa_merged(equal_reads, all_reads, work_dir, outfolder, batch_id, max_seqs_to_spoa, iso_abundance,merged_dict,merged_consensuses,called_consensuses,consensus_map)
612 |     #else:
613 |     #    generate_isoform_using_spoa(equal_reads, all_reads, work_dir, outfolder, batch_id, max_seqs_to_spoa, iso_abundance)
614 | 
615 | 
616 | if __name__ == "__main__":
617 |     main()
618 |     """
619 |     main concept: 
620 |     while reads_for_isoforms:
621 |     start at s and iterate through the nodes up to t.
622 |     Greedy approach:
623 |     get nextnode by getting all out_edges.
624 |     follow the edge which has
625 |     max(current_node['reads']and nextnode['reads'])
626 |     meaning which has the maximum amount of elements which are also in current_node
627 |     all reads that are not in nextnode are deleted from supported_reads
628 |     One isoform:
629 |     Supported_reads as soon as edge[1]==t
630 |     Solve isoform by using spoa.
631 |     """
632 | 


--------------------------------------------------------------------------------
/modules/GraphGeneration.py:
--------------------------------------------------------------------------------
  1 | import tempfile
  2 | import shutil
  3 | import networkx as nx
  4 | import pickle
  5 | import os
  6 | 
  7 | from collections import namedtuple
  8 | from collections import deque
  9 | 
 10 | Read_infos = namedtuple('Read_Infos', 'start_mini_end end_mini_start original_support')
 11 | 
 12 | """IsONform script containing the methods used to generate the Directed Graph from the Intervals coming from the Weighted Interval Scheduling Problem
 13 | Author: Alexander Petri
 14 | The main method in this script was used for debugging therefore is not used during IsONforms actual run.
 15 | """
 16 | 
 17 | 
 18 | def depth_first_search(graph, start):
 19 |     stack = [start]
 20 |     visited = set()
 21 |     while stack:
 22 |         vertex = stack.pop()
 23 |         if vertex in visited:
 24 |             continue
 25 |         yield vertex
 26 |         visited.add(vertex)
 27 |         for neighbor in graph[vertex]:
 28 |             stack.append(neighbor)
 29 | 
 30 | 
 31 | 
 32 | def bfs(DG, startnode): #function for BFS
 33 |     visited = set()
 34 |     queue = deque([startnode])
 35 | 
 36 |     while queue: # Creating loop to visit each node
 37 |         m = queue.popleft()
 38 |         for neighbour in DG.successors(m):
 39 |             if neighbour not in visited:
 40 |                 visited.add(neighbour)
 41 |                 queue.append(neighbour)
 42 |             if neighbour == startnode:
 43 |                 return True
 44 |     return False
 45 | 
 46 | 
 47 | def isCyclicUtil(DG, visited,rec_stack, node):
 48 |     """
 49 |     Used to find out whether adding 'node' led to the graph having a cycle
 50 |     Implemented according to https://www.geeksforgeeks.org/detect-cycle-in-a-graph/
 51 |     """
 52 |     visited.add(node)
 53 |     rec_stack.add(node)
 54 |     # Mark current node as visited and
 55 |     # adds to recursion stack
 56 |     #CNode = namedtuple('CNode', 'visited recStack')
 57 |     #cnode = CNode(True, True)
 58 |     #nodes_dict[node] = cnode
 59 |     # Recur for all neighbours
 60 |     # if any neighbour is visited and in
 61 |     # recStack then graph is cyclic
 62 |     for successor in DG.successors(node):
 63 |         #neighbour = out_edge[1]
 64 |         #if neighbour not in nodes_dict:
 65 |         #    cnode = CNode(False, False)
 66 |         #    nodes_dict[neighbour] = cnode
 67 |         if successor not in visited:
 68 |             if isCyclicUtil(DG, visited, rec_stack, successor):
 69 |                 return True
 70 |         elif successor in rec_stack:
 71 |             return True
 72 |     # The node needs to be popped from
 73 |     # recursion stack before function ends
 74 |     #prev_visited = nodes_dict[node].visited
 75 |     #nodes_dict[node] = CNode(prev_visited, False)
 76 |     rec_stack.remove(node)
 77 |     return False
 78 | 
 79 | 
 80 | def cycle_finder(DG, start_node):
 81 |     """
 82 |     Method used to detect cycles in our graph structure
 83 |     INPUT:
 84 |         DG:         the graph structure
 85 |         start_node: a node that should be located in a cycle
 86 |     OUTPUT:
 87 |         bool:       indicator whether a cycle has been found
 88 |     """
 89 |     #nodes_dict = {}
 90 |     visited=set()
 91 |     rec_stack=set()
 92 |     if isCyclicUtil(DG, visited,rec_stack, start_node):
 93 |         return True
 94 |     return False
 95 | 
 96 | 
 97 | def add_prior_read_infos(inter, r_id, prior_read_infos, name, k):
 98 |     """
 99 |     Function to add read information of the current interval to prior_read_infos, if they do not exist there yet
100 |     INPUT:      inter:              the current interval holding the information
101 |                 r_id:               the read we are looking at
102 |                 prior_read_infos:   information about other reads
103 |                 name:               name of the node we appoint the information to
104 |                 k:                  minimizer length
105 |     OUTPUT:     prior_read_infos:   information about other reads extended by this intervals infos
106 |     """
107 |     read_id = inter[3][slice(0, len(inter[3]),
108 |                              3)]  # recover the read id from the array of instances which was delivered with all_intervals_for_graph
109 |     start_coord = inter[3][slice(1, len(inter[3]),
110 |                                  3)]  # recover the start coordinate of an interval from the array of instances
111 |     end_coord = inter[3][slice(2, len(inter[3]), 3)]
112 |     # iterate through the retreived information and store it in prior_read_infos only for subsequent reads
113 |     for i, r in enumerate(read_id):
114 |         if not r <= r_id:
115 |             start = start_coord[i] + k
116 |             end = end_coord[i]
117 |             tuple_for_data_structure = (r, start, end)
118 |             if tuple_for_data_structure not in prior_read_infos:
119 |                 prior_read_infos[tuple_for_data_structure] = name
120 | 
121 | 
122 | def convert_array_to_hash(info_array):
123 |     """Helper method to convert the array delivered with all_intervals_for_graph into a hash value to more efficiently look up occurence
124 |     This method additionally deletes the first three entries of the array as they contain the infos about this interval occurence, which changes in between instances
125 |     INPUT:  info_array: The array, which was delivered with the interval to indicate where the interval occurs in other reads
126 |     OUTPUT: hash(tup):  A has value of the tuple the shortened interval was converted into. This hash makes it easy to see whether the interval is already present in the read
127 |     """
128 |     # preprocessing: Delete the first three elements from the array, as they contain the information about this occurrence
129 |     for x in range(0, 3):
130 |         if info_array:
131 |             info_array.pop(0)
132 | 
133 |     tup = hash(tuple(info_array))
134 |     return tup
135 | 
136 | 
137 | 
138 | 
139 | def add_edge_support(edge_support, previous_node, name, r_id):
140 |     edge_support[previous_node, name] = []
141 |     edge_support[previous_node, name].append(r_id)
142 | 
143 | 
144 | def known_old_node_action(alternatives_filtered, previous_node, this_len, nodes_for_graph, inter, k, seq, node_sequence,
145 |                           r_id, DG, edge_support, alternative_nodes, old_node, alt_info_tuple, name, alt_cyc_nodes):
146 |     # if we have found a node which this info can be added to
147 |     if alternatives_filtered:
148 |         node_info = alternatives_filtered[0]
149 |         # print("Ninfo",node_info)
150 |         name = node_info[0]
151 |         # update the read information of node name
152 |         prev_nodelist = nodes_for_graph[name]
153 |         r_infos = Read_infos(inter[0], inter[1], True)
154 |         end_mini_seq = seq[inter[1]:inter[1] + k]
155 |         node_sequence[name] = end_mini_seq
156 |         prev_nodelist[r_id] = r_infos
157 |         nodes_for_graph[name] = prev_nodelist
158 |         if not DG.has_edge(previous_node, name):
159 |             DG.add_edge(previous_node, name, this_len)
160 |             #cycle_added2 = cycle_finder(DG, previous_node)
161 |             cycle_added2 = bfs(DG, previous_node)
162 |             if cycle_added2:
163 |                 cycle_added(name, alt_cyc_nodes, inter, DG, previous_node, r_id, seq, node_sequence, k, nodes_for_graph,
164 |                             this_len, edge_support)
165 |             else:
166 |                 add_edge_support(edge_support, previous_node, name, r_id)
167 |         else:
168 |             edge_info = edge_support[previous_node, name]
169 |             if r_id not in edge_info:
170 |                 edge_info.append(r_id)
171 |                 edge_support[previous_node, name] = edge_info
172 |     # if we have not found a node which this info can be added to
173 |     else:
174 |         no_node_to_add_to_action(alternative_nodes, old_node, alt_info_tuple, inter, seq, k, node_sequence, name, r_id,
175 |                                  nodes_for_graph, DG, this_len, previous_node, edge_support)
176 | 
177 | 
178 | 
179 | def new_interval_action(seq, inter, r_id, node_sequence, nodes_for_graph, known_intervals, k, previous_end,
180 |                         prior_read_infos, DG, previous_node, edge_support, name):
181 |     #print("NIA")
182 |     nodelist = {}
183 |     # add the read information for the node
184 |     r_infos = Read_infos(inter[0], inter[1], True)
185 |     end_mini_seq = seq[inter[1]:inter[1] + k]
186 |     node_sequence[name] = end_mini_seq
187 |     nodelist[r_id] = r_infos
188 |     # add a node into nodes_for_graph
189 |     nodes_for_graph[name] = nodelist
190 |     known_intervals[r_id - 1].append((inter[0], name, inter[1]))
191 |     # get the length between the previous end and this nodes start
192 |     length = inter[0] - previous_end
193 |     add_prior_read_infos(inter, r_id, prior_read_infos, name, k)
194 |     DG.add_node(name)
195 |     # connect the node to the previous one
196 |     DG.add_edge(previous_node, name, length=length)
197 |     if DEBUG:
198 |         print("adding edge clear 1035", previous_node, ",", name)
199 |     add_edge_support(edge_support, previous_node, name, r_id)
200 | 
201 | 
202 | def no_node_to_add_to_action(alternative_nodes, old_node, alt_info_tuple, inter, seq, k, node_sequence, name, r_id,
203 |                              nodes_for_graph, DG, this_len, previous_node, edge_support):
204 |     # add a new entry to alternative_nodes[old_node] to enable finding this instance
205 |     alternative_nodes[old_node].append(alt_info_tuple)
206 |     if DEBUG:
207 |         print("oldnodeFurther", old_node)
208 |     # add the read information for the node
209 |     nodelist = {}
210 |     r_infos = Read_infos(inter[0], inter[1], True)
211 |     end_mini_seq = seq[inter[1]:inter[1] + k]
212 |     node_sequence[name] = end_mini_seq
213 |     nodelist[r_id] = r_infos
214 |     nodes_for_graph[name] = nodelist
215 |     DG.add_node(name)
216 |     # get the length between the previous end and this nodes start
217 |     length = this_len
218 |     # connect the node to the previous one
219 |     DG.add_edge(previous_node, name, length=length)
220 |     add_edge_support(edge_support, previous_node, name, r_id)
221 | 
222 | 
223 | def cycle_added(name, alt_cyc_nodes, inter, DG, previous_node, r_id, seq, node_sequence, k, nodes_for_graph, length, edge_support):
224 | 
225 |     old_name = name
226 |     # alt_cyc_nodes is just used to keep track of how many nodes we add over the whole graph generation
227 |     if name not in alt_cyc_nodes.keys():
228 |         alt_cyc_list = []
229 |         alt_cyc_list.append(name)
230 |         alt_cyc_nodes[old_name] = alt_cyc_list
231 |     else:
232 |         alt_cyc_list = alt_cyc_nodes[old_name]
233 |         alt_cyc_list.append(name)
234 |         alt_cyc_nodes[old_name] = alt_cyc_list
235 |     DG.remove_edge(previous_node, name)
236 |     # TODO: add alt_cyc_node instead ofadding a new node to reduce overall nr nodes in our graph
237 | 
238 |     name = str(inter[0]) + ", " + str(inter[1]) + ", " + str(r_id)
239 |     if not DG.has_node(name):
240 |         DG.add_node(name)
241 |     nodelist = {}
242 | 
243 |     r_infos = Read_infos(inter[0], inter[1], True)
244 |     end_mini_seq = seq[inter[1]:inter[1] + k]
245 |     node_sequence[name] = end_mini_seq
246 |     nodelist[r_id] = r_infos
247 |     nodes_for_graph[name] = nodelist
248 |     DG.add_edge(previous_node, name, length=length)
249 |     add_edge_support(edge_support, previous_node, name, r_id)
250 | 
251 | def generateGraphfromIntervals(all_intervals_for_graph, k, delta_len, read_len_dict, all_reads):
252 |     """ generates a networkx graph from the intervals given in all_intervals_for_graph.
253 |     # INPUT:    all_intervals_for_graph:    A dictonary holding lists of minimizer intervals.
254 |                 k:                          K-mer length
255 |                 delta_len:                  integer holding the maximum lenght difference for two reads to still land in the same Isoform
256 |                 readlen_dict:               dictionary holding the read_id as key and the length of the read as value
257 |     #OUTPUT:    result                      a tuple of different output values of the algo
258 |                 DG                          The Networkx Graph object
259 |                 known_intervals             A list of intervals used to check the correctness of the algo
260 |                 reads_for_isoforms          A dictionary holding the reads which are to be put in the same isoform
261 |                 reads_at_start_dict
262 | 
263 |     """
264 | 
265 |     DG = nx.DiGraph()
266 |     # add the read ids to the startend_list
267 |     reads_at_start_dict = {}
268 |     reads_at_end_dict = {}
269 |     reads_for_isoforms = []
270 |     # holds the r_id as key and a list of tuples as value: For identification of reads, also used to ensure correctness of graph
271 |     known_intervals = []
272 |     # the following dictionary is supposed to hold the end minimizer sequence for each node
273 |     node_sequence = {}
274 |     edge_support = {}
275 |     prior_read_infos = {}
276 |     nodes_for_graph = {}
277 |     alternative_nodes = {}
278 |     alt_cyc_nodes={}
279 |     for i in range(1, len(all_intervals_for_graph) + 1):
280 |         reads_at_start_dict[i] = Read_infos(0, 0, True)
281 |         reads_for_isoforms.append(i)
282 |     for i in range(1, len(read_len_dict) + 1):
283 |         reads_at_end_dict[i] = Read_infos(read_len_dict[i], read_len_dict[i], True)
284 |     # a source and a sink node are added to the graph in order to have a well-defined start and end for the paths
285 |     DG.add_node("s", reads=reads_at_start_dict, end_mini_seq='')
286 |     # print("REads at end dict",reads_at_end_dict)
287 |     DG.add_node("t", reads=reads_at_end_dict, end_mini_seq='')
288 |     # adds an empty list for each r_id to known_intervals. To those lists, tuples, representing the intervals are added
289 |     # for _ in itertools.repeat(None, len(all_intervals_for_graph)):
290 |     for i in range(len(all_intervals_for_graph)):
291 |         known_intervals.append([])
292 |     # iterate through the different reads stored in all_intervals_for_graph. For each read one path is built up from source to sink if the nodes needed for that are not already present
293 |     # intervals_for_read holds all intervals which make up the solution for the WIS of a read
294 |     for r_id, intervals_for_read in all_intervals_for_graph.items():
295 |         if DEBUG:
296 |             print('CURR READ:', r_id)
297 |             print("NR Nodes", len(DG.nodes()))
298 |             print("NR Edges", len(DG.edges()))
299 |         containscycle = False
300 |         # set previous_node to be s. This is the node all subsequent nodes are going to have edges with
301 |         previous_node = "s"
302 |         previous_end = 0
303 |         # read_hashs is a dictionary storing hash values as keys and the respective node ids as values
304 |         read_hashs = {}
305 |         # the name of each node is defined to be readID, startminimizerpos , endminimizerpos
306 |         # iterate over all intervals, which are in the solution of a certain read
307 |         for pos, inter in enumerate(intervals_for_read):
308 |             if DEBUG:
309 |                 print("PRNode", previous_node)
310 |             info_tuple = (r_id, inter[0], inter[1])
311 |             if DEBUG:
312 |                 print("INter", inter[0])
313 |                 print(info_tuple)
314 |             # generate hash value of the intervals' infos
315 |             curr_hash = convert_array_to_hash(inter[3])
316 |             if curr_hash in read_hashs:
317 |                 is_repetative = True
318 |                 #cycle_start = read_hashs[curr_hash][-1]
319 |             else:
320 |                 is_repetative = False
321 |             # access prior_read_infos, if the same interval was already found in previous reads
322 |             if info_tuple in prior_read_infos:
323 |                 if DEBUG:
324 |                     print("in prior read infos")
325 |                 # if the interval repeats during this read
326 |                 if is_repetative:
327 |                     if DEBUG:
328 |                         print("cycle not close enough")
329 |                     nodelist = {}
330 |                     this_len = inter[0] - previous_end
331 |                     # add a node into nodes_for_graph
332 |                     name = str(inter[0]) + ", " + str(inter[1]) + ", " + str(r_id)
333 |                     seq = all_reads[r_id][1]
334 |                     end_mini_seq = seq[inter[1]:inter[1] + k]
335 |                     node_sequence[name] = end_mini_seq
336 |                     r_infos = Read_infos(inter[0], inter[1], True)
337 |                     nodelist[r_id] = r_infos
338 |                     nodes_for_graph[name] = nodelist
339 |                     DG.add_node(name)
340 |                     # get the length between the previous end and this nodes start
341 |                     length = this_len
342 |                     # connect the node to the previous one
343 |                     if DEBUG:
344 |                         print("Adding edge, newnode no similar cycles connecting prevnode, name")
345 |                     DG.add_edge(previous_node, name, length=length)
346 |                     if DEBUG:
347 |                         print("adding edge clear 553", previous_node, ",", name)
348 |                     add_edge_support(edge_support, previous_node, name, r_id)
349 |                 # the interval did not repeat during this read
350 |                 else:
351 |                     if DEBUG:
352 |                         print("non repetative interval")
353 |                     # get the name by accessing prior_read_infos
354 |                     name = prior_read_infos[info_tuple]
355 |                     #print("NAME",name)
356 |                     # get length from previous_end to this_start
357 |                     this_len = inter[0] - previous_end
358 |                     # if there is not yet an edge connecting previous_node and name: add edge
359 |                     if not DG.has_edge(previous_node, name):
360 |                         if DEBUG:
361 |                             print("no edge between ", previous_node, " and ", name)
362 |                         seq = all_reads[r_id][1]
363 |                         # update the read information of node name
364 |                         prev_nodelist = nodes_for_graph[name]
365 |                         r_infos = Read_infos(inter[0], inter[1], True)
366 |                         end_mini_seq = seq[inter[1]:inter[1] + k]
367 |                         node_sequence[name] = end_mini_seq
368 |                         prev_nodelist[r_id] = r_infos
369 |                         nodes_for_graph[name] = prev_nodelist
370 |                         length = this_len
371 |                         #is_cyclic = SimplifyGraph.isCyclic(DG)
372 |                         #print("SimplifyGraph_cyclic ", is_cyclic)
373 |                         DG.add_edge(previous_node, name, length=length)
374 |                         #print("edge ", previous_node, " to ", name, "added")
375 |                         #cycle_added2 = cycle_finder(DG, previous_node)
376 |                         cycle_added2 = bfs(DG,previous_node)
377 |                         #is_cyclic = SimplifyGraph.isCyclic(DG)
378 |                         #try:
379 |                         #    nx_cycle=nx.find_cycle(DG)
380 |                         #    print("Networkx found a cycle ",nx_cycle)
381 |                         #except:
382 |                         #    print("no cycle was found by networkx")
383 |                         #if is_cyclic:
384 |                             #    k_size+=1
385 |                             #    w+=1
386 |                             #print("The graph has a cycle - critical error")
387 |                         #if is_cyclic != cycle_added2:
388 |                         #    print("SimplifyGraph_cyclic ",is_cyclic)
389 |                         #    print("GG_cyclic ", cycle_added2, ", ",previous_node)
390 |                         #    print("ERROR- different outcome")
391 |                         #the edge we added did introduce a cycle in our graph. We remove the edge again and instead generate a new node
392 |                         #if cycle_added2:
393 |                         if cycle_added2:
394 |                             cycle_added(name, alt_cyc_nodes, inter, DG, previous_node, r_id, seq, node_sequence, k, nodes_for_graph, length, edge_support)
395 |                         else:
396 |                             add_edge_support(edge_support, previous_node, name, r_id)
397 |                     # if there is an edge connecting previous_node and name: test if length difference is not higher than delta_len
398 |                     else:
399 |                         if DEBUG:
400 |                             print("Has edge")
401 |                             print(previous_node, name)
402 |                         prev_len = DG[previous_node][name]["length"]
403 |                         len_difference = abs(this_len - prev_len)
404 |                         # if the length difference is <delta_len: Just add the readinfos
405 |                         if len_difference < delta_len:
406 |                             # update the read information of node name
407 |                             prev_nodelist = nodes_for_graph[name]
408 |                             seq = all_reads[r_id][1]
409 |                             r_infos = Read_infos(inter[0], inter[1], True)
410 |                             end_mini_seq = seq[inter[1]:inter[1] + k]
411 |                             node_sequence[name] = end_mini_seq
412 |                             prev_nodelist[r_id] = r_infos
413 |                             nodes_for_graph[name] = prev_nodelist
414 |                             edge_info = edge_support[previous_node, name]
415 |                             if r_id not in edge_info:
416 |                                 edge_info.append(r_id)
417 |                                 edge_support[previous_node, name] = edge_info
418 |                         # if the length difference is >delta len: generate new node, to be able to tell those nodes apart we use alternative_nodes
419 |                         else:
420 |                             #print("LEN diff > DL")
421 |                             nodelist = {}
422 |                             # inappropriate_node
423 |                             old_node = name
424 |                             name = str(inter[0]) + ", " + str(inter[1]) + ", " + str(r_id)
425 |                             alt_info_tuple = (name, previous_node, prev_len)
426 |                             # add old node as key for alternative nodes: This is to find the alternative nodes for this one easily
427 |                             if not (old_node in alternative_nodes):
428 |                                 alternative_nodes[old_node] = []
429 |                                 DG.add_node(name)
430 |                                 r_infos = Read_infos(inter[0], inter[1], True)
431 |                                 end_mini_seq = seq[inter[1]:inter[1] + k]
432 |                                 node_sequence[name] = end_mini_seq
433 |                                 nodelist[r_id] = r_infos
434 |                                 nodes_for_graph[name] = nodelist
435 |                                 DG.add_edge(previous_node, name, length=this_len)
436 |                                 add_edge_support(edge_support, previous_node, name, r_id)
437 |                             # if we already know old_node (it is a key in alternative_nodes)
438 |                             else:
439 |                                 alternative_infos_list = alternative_nodes[old_node]
440 |                                 alternatives_filtered = [item for item in alternative_infos_list if
441 |                                                          previous_node == item[1] and abs(
442 |                                                              this_len - item[2]) < delta_len]
443 |                                 known_old_node_action(alternatives_filtered, previous_node, this_len,
444 |                                                       nodes_for_graph, inter, k,
445 |                                                       seq, node_sequence, r_id, DG, edge_support, alternative_nodes,
446 |                                                       old_node,
447 |                                                       alt_info_tuple, name, alt_cyc_nodes)
448 | 
449 |                 # keep known_intervals up to date
450 |                 known_intervals[r_id - 1].append((inter[0], name, inter[1]))
451 |             # if the information for the interval was not yet found in a previous read (meaning the interval is new)
452 |             else:
453 |                 seq = all_reads[r_id][1]
454 |                 name = str(inter[0]) + ", " + str(inter[1]) + ", " + str(r_id)
455 |                 new_interval_action(seq, inter, r_id, node_sequence, nodes_for_graph, known_intervals, k, previous_end,
456 |                                     prior_read_infos, DG, previous_node, edge_support, name)
457 | 
458 |             # set the previous node for the next iteration
459 |             previous_node = name
460 |             if DEBUG:
461 |                 print("PREVNODE", previous_node)
462 |             previous_end = inter[1]
463 |             # find out whether the current hash has already been added to read_hashs
464 |             if not is_repetative:
465 |                 # if the current hash was not yet present in read_hashs: Add it
466 |                 read_hashs[curr_hash] = []
467 |                 read_hashs[curr_hash].append(name)
468 |         # add an edge from name to "t" as name was the last node in the read
469 |         if not DG.has_edge(name, "t"):
470 |             DG.add_edge(name, "t", length=0)
471 |             if DEBUG:
472 |                 print("adding edge clear 1066", name, ",", "t")
473 |             edge_support[name, "t"] = []
474 |             edge_support[name, "t"].append(r_id)
475 |         else:
476 |             edge_info = edge_support[name, "t"]
477 |             if r_id not in edge_info:
478 |                 edge_info.append(r_id)
479 |                 edge_support[name, "t"] = edge_info
480 |     print("Number of alternative nodes due to cycle",len(alt_cyc_nodes.keys()))
481 |     # set the node attributes to be nodes_for_graph, very convenient way of solving this
482 |     nx.set_node_attributes(DG, nodes_for_graph, name="reads")
483 |     nx.set_node_attributes(DG, node_sequence, name="end_mini_seq")
484 |     nx.set_edge_attributes(DG, edge_support, name='edge_supp')
485 | 
486 |     # return DG and known_intervals, used to
487 |     result = (DG, reads_for_isoforms)
488 |     return result
489 | 
490 | 
491 | """
492 | Plots the given Directed Graph via Matplotlib
493 | Input: DG   Directed Graph
494 | """
495 | 
496 | 
497 | 
498 | 
499 | 
500 | def get_read_lengths(all_reads):
501 |     """Helper method which extracts the read lengths from all_reads. We will use those during the graph generation to appoint more meaningful information to the node't'
502 |     INPUT: all_reads: dictionary which holds the overall read infos key: r_id, value tuple(readname, sequence, some info i currently don't care about)
503 |     OUTPUT: readlen_dict: dictionary holding the read_id as key and the length of the read as value
504 |     """
505 |     readlen_dict = {}
506 |     for r_id, infos in all_reads.items():
507 |         seq = infos[1]
508 |         seqlen = len(seq)
509 | 
510 |         readlen_dict[r_id] = seqlen
511 |     return readlen_dict
512 | 
513 | 
514 | """
515 | USED FOR DEBUGGING ONLY-deprecated in IsONform
516 | """
517 | 
518 | DEBUG=False
519 | def main():
520 |     # import sys
521 |     # sys.stdout = open('log.txt', 'w')
522 |     print('test')
523 |     w = 10
524 |     reads = 62
525 |     max_seqs_to_spoa = 200
526 |     work_dir = tempfile.mkdtemp()
527 |     # print("Temporary workdirektory:", work_dir)
528 |     k_size = 9
529 |     outfolder = "out_local"
530 |     cwd = os.getcwd()
531 |     # print("CWD",cwd)
532 |     file = open('all_intervals.txt', 'rb')
533 |     all_intervals_for_graph = pickle.load(file)
534 |     file.close()
535 |     # print("All of them", len(all_intervals_for_graph))
536 | 
537 |     file2 = open('all_reads.txt', 'rb')
538 | 
539 |     all_reads = pickle.load(file2)
540 |     print("Allreads type")
541 |     # print(type(all_reads))
542 |     # print(all_reads)
543 |     file2.close()
544 |     delta_len = 2 * k_size
545 |     # is_cyclic = True
546 |     # while is_cyclic:
547 |     read_len_dict = get_read_lengths(all_reads)
548 |     print("Generating graph")
549 |     DG, known_intervals, reads_for_isoforms, reads_list = generateGraphfromIntervals(
550 |         all_intervals_for_graph, k_size, delta_len, read_len_dict, all_reads)
551 | 
552 |     print("edges with attributes:")
553 | 
554 |     print("Calling the method")
555 | 
556 |     # simplifyGraph(DG, all_reads,work_dir,k_size,delta_len,known_intervals)
557 |     print("Graph simplified")
558 |     # simplifyGraphOriginal(DG,delta_len, all_reads, work_dir, k_size,known_intervals)
559 |     # print("#Nodes for DG: " + str(DG.number_of_nodes()) + " , #Edges for DG: " + str(DG.number_of_edges()))
560 |     # draw_Graph(DG)
561 |     # print(DG.nodes["s"]["reads"])
562 |     # print(list(DG.nodes(data=True)))
563 |     # print("edges with attributes:")
564 |     # print(DG.edges(data=True))
565 |     # print("ReadNodes")
566 |     # print("all edges for the graph")
567 |     # print([e for e in DG.edges])
568 |     # draw_Graph(DG)
569 |     # print("knownintervals")
570 |     # print(known_intervals)
571 |     # The call for the isoform generation (deprecated during implementation)
572 |     # TODO: invoke isoform_generation again
573 |     # generate_isoforms(DG, all_reads, reads_for_isoforms, work_dir, outfolder, max_seqs_to_spoa)
574 |     # DG.nodes(data=True)
575 | 
576 |     # print("removing temporary workdir")
577 |     shutil.rmtree(work_dir)
578 | 
579 | 
580 | if __name__ == "__main__":
581 |     main()
582 | 


--------------------------------------------------------------------------------
/main:
--------------------------------------------------------------------------------
  1 | #! /usr/bin/env python
  2 | from __future__ import print_function
  3 | import argparse
  4 | from array import array
  5 | import itertools
  6 | import os
  7 | import shutil
  8 | import sys
  9 | import tempfile
 10 | import pickle
 11 | import operator
 12 | from collections import defaultdict, deque
 13 | 
 14 | 
 15 | from modules import help_functions, GraphGeneration, batch_merging_parallel, IsoformGeneration, SimplifyGraph
 16 | 
 17 | D = {chr(i) : min( 10**( - (ord(chr(i)) - 33)/10.0 ), 0.79433)  for i in range(128)}
 18 | 
 19 | 
 20 | def write_batch(reads,outfolder,batch_pickle):
 21 |     this_batch_dict = {}
 22 |     for id, (acc, seq, qual) in reads.items():
 23 |         this_batch_dict[acc] = seq
 24 |     pickle_batch_file = open(os.path.join(outfolder, batch_pickle), 'wb')
 25 |     pickle.dump(this_batch_dict, pickle_batch_file)
 26 |     pickle_batch_file.close()
 27 | 
 28 | 
 29 | 
 30 | def get_read_lengths(all_reads):
 31 |     """Helper method which extracts the read lengths from all_reads. We will use those during the graph generation to appoint more meaningful information to the node't'
 32 |         INPUT: all_reads: dictionary which holds the overall read infos key: r_id, value tuple(readname, sequence, some info i currently don't care about)
 33 |         OUTPUT: readlen_dict: dictionary holding the read_id as key and the length of the read as value
 34 |     """
 35 |     readlen_dict = {}
 36 |     for r_id, infos in all_reads.items():
 37 |         seq = infos[1]
 38 |         seqlen = len(seq)
 39 |         readlen_dict[r_id] = seqlen
 40 |     return readlen_dict
 41 | 
 42 | 
 43 | def eprint(*args, **kwargs):
 44 |     print(*args, file=sys.stderr, **kwargs)
 45 | 
 46 | 
 47 | def remove_read_polyA_ends(seq, threshold_len, to_len):
 48 |     """
 49 |     Funtion that removes the polyA_ends from the reads. We remove all polyAtails longer than threshold_len by transforming them into polyA strings of length to_len.
 50 |     INPUT:  seq:            the sequence to be altered
 51 |             threshold_len:  the length threshold over which we alter the polyA sequences
 52 |             to_len:         the length of the poly_A tails after the alteration
 53 |     OUTPUT: seq_mod:        the sequence that has been modified by the function, i.e. the sequence with shortened polyA tails
 54 |     """
 55 |     #we only want to alter polyA sequences that are located in the end of the read->calculate a window length in which we perform the change
 56 |     end_length_window = min(len(seq)//2, 100)
 57 |     seq_list = [ seq[:-end_length_window] ]
 58 | 
 59 |     for ch, g in itertools.groupby(seq[-end_length_window:]):
 60 |         h_len = sum(1 for x in g)
 61 |         if h_len > threshold_len and (ch == "A" or ch == "T"):
 62 |             seq_list.append(ch*to_len)
 63 |         else:
 64 |             seq_list.append(ch*h_len)
 65 | 
 66 |     seq_mod = "".join([s for s in seq_list])
 67 |     return seq_mod
 68 | 
 69 | 
 70 | 
 71 | 
 72 | def rindex(lst, value):
 73 |     """
 74 |         Function that calculates the reverse index. We want to find the last (but still) smallest k-mer in the window not the first
 75 |         INPUT:  lst:            the window of kmers as a list
 76 |                 value:          the smallest kmer
 77 |         OUTPUT: minimizer_pos:  the last position of kmer with value in lst (not the first as before)
 78 |         """
 79 |     return len(lst) - operator.indexOf(reversed(lst), value) - 1
 80 | 
 81 | def get_kmer_minimizers(seq, k_size, w_size):
 82 |     # kmers = [seq[i:i+k_size] for i in range(len(seq)-k_size) ]
 83 |     w = w_size - k_size
 84 |     #save the window as a deque and instead of the sequence itself we use its hash value
 85 |     window_kmers = deque([hash(seq[i:i+k_size]) for i in range(w +1)])
 86 |     #the smallest kmer (or to be precise the smallest hash) we have found in the window
 87 |     curr_min = min(window_kmers)
 88 |     #we now want the last occurrence of the smallest kmer not the first anymore
 89 |     minimizer_pos = rindex(list(window_kmers), curr_min)
 90 |     #add the initial minimizer to minimizers
 91 |     minimizers = [ (seq[minimizer_pos: minimizer_pos+k_size], minimizer_pos) ] # get the last element if ties in window
 92 |     #iterate over the remaining read and find all minimizers therein
 93 |     for i in range(w+1, len(seq) - k_size):
 94 |         new_kmer = hash(seq[i:i+k_size])
 95 |         # updating window
 96 |         discarded_kmer = window_kmers.popleft()
 97 |         window_kmers.append(new_kmer)
 98 | 
 99 |         # we have discarded previous window's minimizer, look for new minimizer brute force
100 |         if curr_min == discarded_kmer and minimizer_pos < i - w:
101 |             curr_min = min(window_kmers)
102 |             minimizer_pos = rindex(list(window_kmers), curr_min) + i - w
103 |             minimizers.append( (seq[minimizer_pos: minimizer_pos+k_size], minimizer_pos) ) # get the last element if ties in window
104 | 
105 |         # Previous minimizer still in window, we only need to compare with the recently added kmer
106 |         elif new_kmer < curr_min:
107 |             curr_min = new_kmer
108 |             minimizers.append( (seq[i: i+k_size], i) )
109 | 
110 |     return minimizers
111 | 
112 | def get_kmer_maximizers(seq, k_size, w_size):
113 |     # kmers = [seq[i:i+k_size] for i in range(len(seq)-k_size) ]
114 |     w = w_size - k_size
115 |     window_kmers = deque([seq[i:i+k_size] for i in range(w +1)])
116 |     curr_min = max(window_kmers)
117 |     minimizers = [ (curr_min, list(window_kmers).index(curr_min)) ]
118 | 
119 |     for i in range(w+1,len(seq) - k_size):
120 |         new_kmer = seq[i:i+k_size]
121 |         # updating window
122 |         discarded_kmer = window_kmers.popleft()
123 |         window_kmers.append(new_kmer)
124 | 
125 |         # we have discarded previous windows minimizer, look for new minimizer brute force
126 |         if curr_min == discarded_kmer:
127 |             curr_min = max(window_kmers)
128 |             minimizers.append( (curr_min, list(window_kmers).index(curr_min) + i - w ) )
129 | 
130 |         # Previous minimizer still in window, we only need to compare with the recently added kmer
131 |         elif new_kmer > curr_min:
132 |             curr_min = new_kmer
133 |             minimizers.append( (curr_min, i) )
134 | 
135 |     return minimizers
136 | 
137 | 
138 | def get_minimizers_and_positions_compressed(reads, w, k, hash_fcn):
139 |     # 1. homopolymenr compress read and obtain minimizers
140 |     M = {}
141 |     for r_id in reads:
142 |         (acc, seq, qual) = reads[r_id]
143 | 
144 |         seq_hpol_comp = ''.join(ch for ch, _ in itertools.groupby(seq))
145 | 
146 |         if hash_fcn == "lex":
147 |             minimizers = get_kmer_minimizers(seq_hpol_comp, k, w)
148 |         elif hash_fcn == "rev_lex":
149 |             minimizers = get_kmer_maximizers(seq_hpol_comp, k, w)
150 |         # indicies we want to take quality values from to get quality string of homopolymer compressed read
151 |         indices = [i for i, (n1,n2) in enumerate(zip(seq[:-1],seq[1:])) if n1 != n2]
152 |         indices.append(len(seq) - 1)
153 |         positions_in_non_compressed_sring = [(m, indices[p]) for m, p in minimizers ]
154 |         M[r_id] = positions_in_non_compressed_sring
155 | 
156 |     return M
157 | 
158 | 
159 | def get_minimizers_and_positions(reads, w, k, hash_fcn):
160 |     # 1. homopolymenr compress read and obtain minimizers
161 |     M = {}
162 |     for r_id in reads:
163 |         (acc, seq, qual) = reads[r_id]
164 |         if hash_fcn == "lex":
165 |             minimizers = get_kmer_minimizers(seq, k, w)
166 |         elif hash_fcn == "rev_lex":
167 |             minimizers = get_kmer_maximizers(seq, k, w)
168 | 
169 |         M[r_id] = minimizers
170 | 
171 |     return M
172 | 
173 | 
174 | def get_minimizer_combinations_database(M, k, x_low, x_high,reads):
175 |     M2 = defaultdict(lambda: defaultdict(lambda: array("I")))
176 |     tmp_cnt = 0
177 |     forbidden = 'A'*k
178 |     for r_id in M:
179 |         minimizers = M[r_id]
180 |         for (m1,p1), m1_curr_spans in minimizers_comb_iterator(minimizers, k, x_low, x_high):
181 |             for (m2, p2) in m1_curr_spans:
182 |                 if m2 == m1 == forbidden:
183 |                     continue
184 | 
185 |                 tmp_cnt +=1
186 |                 M2[m1][m2].append(r_id)
187 |                 M2[m1][m2].append(p1)
188 |                 M2[m1][m2].append(p2)
189 | 
190 |     print(tmp_cnt, "MINIMIZER COMBINATIONS GENERATED")
191 | 
192 |     avg_bundance = 0
193 |     singleton_minimzer = 0
194 |     cnt = 1
195 |     for m1 in list(M2.keys()):
196 |         for m2 in list(M2[m1].keys()):
197 |             #if the minimizer_pair is represented at more than one occurrence
198 |             if len(M2[m1][m2]) > 3:
199 |                 avg_bundance += len(M2[m1][m2])//3
200 |                 cnt += 1
201 |             #the minimizer_combination is only once in the data therefore does not yield viable information for the graph later on
202 |             else:
203 |                 del M2[m1][m2]
204 |                 singleton_minimzer += 1
205 |             #we also want to filter out minimizer combinations if they are too abundant (more than 3 times per read)
206 |             if len(M2[m1][m2]) // 3 > 3 * len(reads):
207 |                 del M2[m1][m2]
208 | 
209 |     print("Average abundance for non-unique minimizer-combs:", avg_bundance/float(cnt))
210 |     print("Number of singleton minimizer combinations filtered out:", singleton_minimzer)
211 | 
212 |     return M2
213 | 
214 | 
215 | def minimizers_comb_iterator(minimizers, k, x_low, x_high):
216 |     for i, (m1, p1) in enumerate(minimizers[:-1]):
217 |         m1_curr_spans = []
218 |         for j, (m2, p2) in enumerate(minimizers[i+1:]):
219 |             if x_low < p2 - p1 and p2 - p1 <= x_high:
220 |                 m1_curr_spans.append((m2, p2))
221 |                 # yield (m1,p1), (m2, p2)
222 |             elif p2 - p1 > x_high:
223 |                 break
224 |         yield (m1, p1), m1_curr_spans[::-1]
225 | 
226 | 
227 | def fill_p2(p, all_intervals_sorted_by_finish):
228 |     stop_to_max_j = {stop: j for j, (start, stop, w, _) in enumerate(all_intervals_sorted_by_finish) if start < stop}
229 |     all_choord_to_max_j = []
230 |     j_max = 0
231 |     for i in range(0, all_intervals_sorted_by_finish[-1][1] + 1):
232 |         if i in stop_to_max_j:
233 |             j_max = stop_to_max_j[i]
234 | 
235 |         all_choord_to_max_j.append(j_max)
236 |     for j, (start, stop, w, _) in enumerate(all_intervals_sorted_by_finish):
237 |         j_max = all_choord_to_max_j[start]
238 |         p.append(j_max)
239 |     return p
240 | 
241 | 
242 | def solve_WIS(all_intervals_sorted_by_finish):
243 |     # Using notation from https://courses.cs.washington.edu/courses/cse521/13wi/slides/06dp-sched.pdf
244 |     p = [None]
245 |     fill_p2(p, all_intervals_sorted_by_finish)
246 |     epsilon = 0.0001
247 |     # w - 1 since the read interval itself is included in the instance
248 |     v = [None] + [(w - 1)*(stop-start + epsilon) for (start, stop, w, _) in all_intervals_sorted_by_finish]
249 |     OPT = [0]
250 |     for j in range(1, len(all_intervals_sorted_by_finish) +1):
251 |         OPT.append( max(v[j] + OPT[ p[j] ], OPT[j-1] ) )
252 |     # Find solution
253 |     opt_indicies = []
254 |     j = len(all_intervals_sorted_by_finish)
255 |     while j >= 0:
256 |         if j == 0:
257 |             break
258 |         if v[j] + OPT[p[j]] > OPT[j-1]:
259 |             opt_indicies.append(j - 1) # we have shifted all indices forward by one so we neew to reduce to j -1 because of indexing in python works
260 |             j = p[j]
261 |         else:
262 |             j -= 1
263 |     return opt_indicies
264 | 
265 | 
266 | def get_intervals_to_correct(opt_indicies, all_intervals_sorted_by_finish):
267 |     intervals_to_correct = []
268 |     for j in opt_indicies:
269 |         start, stop, weights, instance = all_intervals_sorted_by_finish[j]
270 |         intervals_to_correct.append((start, stop, weights, instance))
271 | 
272 |     return intervals_to_correct
273 | 
274 | 
275 | def batch(dictionary, size):
276 |     batches = []
277 |     sub_dict = {}
278 |     for i, (acc, seq) in enumerate(dictionary.items()):
279 |         if i > 0 and i % size == 0:
280 |             batches.append(sub_dict)
281 |             sub_dict = {}
282 |             sub_dict[acc] = seq
283 |         else:
284 |             sub_dict[acc] = seq
285 | 
286 |     if i / size != 0:
287 |         sub_dict[acc] = seq
288 |         batches.append(sub_dict)
289 |     elif len(dictionary) == 1:
290 |         batches.append(sub_dict)
291 | 
292 |     return batches
293 | 
294 | 
295 | def grouper(iterable, n, fillvalue=None):
296 |     "Collect data into fixed-length chunks or blocks"
297 |     # grouper('ABCDEFG', 3, 'x') --> ABC DEF Gxx"
298 |     args = [iter(iterable)] * n
299 |     return itertools.zip_longest(*args, fillvalue=fillvalue)
300 | 
301 | 
302 | def add_items(seqs, r_id, p1, p2):
303 |     seqs.append(r_id)
304 |     seqs.append(p1)
305 |     seqs.append(p2)
306 | 
307 | 
308 | def find_most_supported_span(r_id, m1, p1, m1_curr_spans, minimizer_combinations_database, all_intervals, k_size, delta_len):
309 |     """
310 |     Funtion that detects the most supported spans in the database and adds them to all_intervals.
311 |     INPUT:  r_id:           the id of the read we are workin on
312 |             m1:             the length threshold over which we alter the polyA sequences
313 |             p1:             the length of the poly_A tails after the alteration
314 |             m1_curr_spans:
315 |             minimizer_combinations_database:
316 |             reads:
317 |             all_intervals:  list holding the interval information (empty at this point)
318 |             k_size:
319 |             delta_len:
320 | 
321 |     OUTPUT: all_intervals:     modified list of all intervals
322 |     """
323 |     #print("Most_supp_span")
324 |     for (m2, p2) in m1_curr_spans:
325 |         relevant_reads = minimizer_combinations_database[m1][m2]
326 |         if len(relevant_reads)//3 >= 3:
327 |             seqs = array("I")
328 |             seqs.append(r_id)
329 |             seqs.append(p1)
330 |             seqs.append(p2)
331 |             for relevant_read_id, pos1, pos2 in grouper(relevant_reads, 3): #relevant_reads:
332 |                 if r_id == relevant_read_id:
333 |                     continue
334 |                 elif abs((p2-p1)-(pos2-pos1)) < delta_len:
335 |                     seqs.append(relevant_read_id)
336 |                     seqs.append(pos1)
337 |                     seqs.append(pos2)
338 |             all_intervals.append((p1 + k_size, p2,  len(seqs)//3, seqs))
339 | 
340 | 
341 | def main(args):
342 |     #Todo: add timestamp
343 |     print("ARGS",args)
344 |     all_batch_reads_dict={}
345 |     # start = time()
346 |     if os.path.exists("mapping.txt"):
347 |         os.remove("mapping.txt")
348 |     outfolder = args.outfolder
349 |     #sys.stdout = open(os.path.join(outfolder,"stdout.txt"), "w")
350 |     # read the file and filter out polyA_ends(via remove_read_polyA_ends)
351 |     all_reads = {i + 1: (acc, remove_read_polyA_ends(seq, 12, 1), qual) for i, (acc, (seq, qual)) in enumerate(help_functions.readfq(open(args.fastq, 'r')))}
352 |     max_seqs_to_spoa = args.max_seqs_to_spoa
353 |     if len(all_reads) <= args.exact_instance_limit:
354 |         args.exact = True
355 |     if args.set_w_dynamically:
356 |         args.w = args.k + min(7, int(len(all_reads) / 500))
357 |     delta_iso_len_3 = args.delta_iso_len_3
358 |     delta_iso_len_5 = args.delta_iso_len_5
359 |     work_dir = tempfile.mkdtemp()
360 |     print("Temporary workdirektory:", work_dir)
361 | 
362 |     k_size = args.k
363 |     x_high = args.xmax
364 |     x_low = args.xmin
365 |     if args.parallel:
366 |         filename = args.fastq.split("/")[-1]
367 |         tmp_filename = filename.split("_")
368 |         tmp_lastpart = tmp_filename[-1].split(".")
369 |         p_batch_id = tmp_lastpart[0]
370 |         skipfilename = "skip"+p_batch_id+".fa"
371 | 
372 |     for batch_id, reads in enumerate(batch(all_reads, args.max_seqs)):
373 | 
374 | 
375 |         new_all_reads = {}
376 |         if args.parallel:
377 |             batch_pickle = str(p_batch_id) + "_batch"
378 |         else:
379 |             skipfilename="skip"+str(batch_id)+".fa"
380 |             batch_pickle = str(batch_id) + "batch"
381 |             skipfilename = "skip" + str(batch_id) + ".fa"
382 |         skipfile = open(os.path.join(outfolder, skipfilename), 'w')
383 |         write_batch(reads, outfolder, batch_pickle)
384 |         skipfile=open(os.path.join(outfolder,skipfilename),'w')
385 | 
386 |         if args.set_w_dynamically:
387 |             # Activates for 'small' clusters with less than 700 reads
388 |             if len(reads) >= 100:
389 |                 w = min(args.w, args.k + (
390 |                             len(reads) // 100 + 4))
391 |             elif len(reads) == 1:
392 |                 for id, vals in reads.items():
393 |                     (acc, seq, qual) = vals
394 |                     skipfile.write(">{0}\n{1}\n".format(acc, seq))
395 |                 print("Not enough reads to work on!")
396 |                 continue
397 |             else:
398 |                 w = args.k + 1 + len(reads) // 30
399 |         else:
400 |             w = args.w
401 |         #print("Window used for batch:", w)
402 |         iso_abundance = args.iso_abundance
403 |         delta_len = args.delta_len
404 |         graph_id = 1
405 |         print("Working on {0} reads in a batch".format(len(reads)))
406 |         hash_fcn = "lex"
407 |         not_used=0
408 |         #generate all minimizer combinations
409 |         if args.compression:
410 |             minimizer_database = get_minimizers_and_positions_compressed(reads, w, k_size, hash_fcn)
411 |         else:
412 |             minimizer_database = get_minimizers_and_positions(reads, w, k_size, hash_fcn)
413 | 
414 |         minimizer_combinations_database = get_minimizer_combinations_database(minimizer_database, k_size, x_low,
415 |                                                                               x_high, reads)
416 |         previously_corrected_regions = defaultdict(list)
417 |         all_intervals_for_graph = {}
418 |         for r_id in sorted(reads):
419 |             print(r_id)
420 |             read_min_comb = [((m1, p1), m1_curr_spans) for (m1, p1), m1_curr_spans in
421 |                              minimizers_comb_iterator(minimizer_database[r_id], k_size, x_low, x_high)]
422 | 
423 |             if args.exact:
424 |                 previously_corrected_regions = defaultdict(list)
425 | 
426 |             read_previously_considered_positions = set(
427 |                 [tmp_pos for tmp_p1, tmp_p2, w_tmp, _ in previously_corrected_regions[r_id] for tmp_pos in
428 |                  range(tmp_p1, tmp_p2)])
429 | 
430 |             if args.verbose:
431 |                 if read_previously_considered_positions:
432 |                     eprint("not corrected:", [(p1_, p2_) for p1_, p2_ in
433 |                                               zip(sorted(read_previously_considered_positions)[:-1],
434 |                                                   sorted(read_previously_considered_positions)[1:]) if
435 |                                               p2_ > p1_ + 1])
436 |                 else:
437 |                     eprint("not corrected: entire read", )
438 | 
439 |             if previously_corrected_regions[r_id]:
440 |                 read_previously_considered_positions = set(
441 |                     [tmp_pos for tmp_p1, tmp_p2, w_tmp, _ in previously_corrected_regions[r_id] for tmp_pos in
442 |                      range(tmp_p1, tmp_p2)])
443 |                 group_id = 0
444 |                 pos_group = {}
445 | 
446 |                 if len(read_previously_considered_positions) > 1:
447 |                     sorted_corr_pos = sorted(read_previously_considered_positions)
448 | 
449 |                     for p1, p2 in zip(sorted_corr_pos[:-1], sorted_corr_pos[1:]):
450 | 
451 |                         if p2 > p1 + 1:
452 |                             pos_group[p1] = group_id
453 |                             group_id += 1
454 |                             pos_group[p2] = group_id
455 |                         else:
456 |                             pos_group[p1] = group_id
457 | 
458 |                     if p2 == p1 + 1:
459 |                         pos_group[p2] = group_id
460 | 
461 |             else:
462 |                 read_previously_considered_positions = set()
463 |                 pos_group = {}
464 |             all_intervals = []
465 |             prev_visited_intervals = []
466 |             #print("it over read_min_comb")
467 |             for (m1, p1), m1_curr_spans in read_min_comb:
468 |                 #print(m1,", ", p1)
469 |                 # If any position is not in range of current corrections: then correct, not just start and stop
470 |                 not_prev_corrected_spans = [(m2, p2) for (m2, p2) in m1_curr_spans if not (
471 |                         p1 + k_size in read_previously_considered_positions and p2 - 1 in read_previously_considered_positions)]
472 |                 set_not_prev = set(not_prev_corrected_spans)
473 |                 not_prev_corrected_spans2 = [(m2, p2) for (m2, p2) in m1_curr_spans if
474 |                                              (m2, p2) not in set_not_prev and (
475 |                                                      p1 + k_size in pos_group and p2 - 1 in pos_group and pos_group[
476 |                                                  p1 + k_size] != pos_group[p2 - 1])]
477 |                 not_prev_corrected_spans += not_prev_corrected_spans2
478 | 
479 |                 if not_prev_corrected_spans:  # p1 + k_size not in read_previously_considered_positions:
480 |                     find_most_supported_span(r_id, m1, p1, m1_curr_spans, minimizer_combinations_database,
481 |                                               all_intervals, k_size, delta_len)
482 |             # add prev_visited_intervals to intervals to consider
483 |             all_intervals.extend(prev_visited_intervals)
484 | 
485 |             if previously_corrected_regions[r_id]:  # add previously corrected regions in to the solver
486 | 
487 |                 all_intervals.extend(previously_corrected_regions[r_id])
488 |                 del previously_corrected_regions[r_id]
489 | 
490 |             if not all_intervals:
491 |                 not_used+=1
492 |                 if DEBUG:
493 |                     eprint("Found no reads to work on")
494 |                 vals=reads[r_id]
495 |                 (acc, seq, qual) = vals
496 |                 skipfile.write(">{0}\n{1}\n".format(acc, seq))
497 |                 continue
498 |             else:
499 |                 all_intervals.sort(key=lambda x: x[1])
500 |                 opt_indicies = solve_WIS(
501 |                     all_intervals)  # solve Weighted Interval Scheduling here to find set of best non overlapping intervals to correct over
502 |                 intervals_to_correct = get_intervals_to_correct(opt_indicies[::-1], all_intervals)
503 |                 #if we have found intervals in our read: add it to all_intervals_for_graph and give it a graph_id (an internal id)
504 |                 if intervals_to_correct:
505 |                     all_intervals_for_graph[graph_id] = intervals_to_correct
506 |                     new_all_reads[graph_id] = reads[r_id]
507 |                     graph_id += 1
508 |                 #the read has no intervals in common with other reads-> we add it into skipfile
509 |                 else:
510 |                     not_used += 1
511 |                     if DEBUG:
512 |                         eprint("Found no reads to work on")
513 |                     vals = reads[r_id]
514 |                     (acc, seq, qual) = vals
515 |                     skipfile.write(">{0}\n{1}\n".format(acc, seq))
516 | 
517 |         if not_used>0:
518 |             print("Skipped ",not_used," reads due to not having high enough interval abundance")
519 |         else:
520 |             print("Working on all reads")
521 |         print("Generating the graph")
522 |         all_batch_reads_dict[batch_id] = new_all_reads
523 |         read_len_dict = get_read_lengths(all_reads)
524 |         #for key,value in all_intervals_for_graph.items():
525 |         #    print(key,len(value))
526 |         #print(all_intervals_for_graph)
527 | 
528 |         #profiler = Profiler()
529 |         #profiler.start()
530 |         #generate the graph from the intervals
531 |         #TODO: add timestamp
532 |         DG,  reads_for_isoforms = GraphGeneration.generateGraphfromIntervals(
533 |             all_intervals_for_graph, k_size, delta_len, read_len_dict,new_all_reads)
534 |         #profiler.stop()
535 | 
536 |         #profiler.print()
537 |         #test for cyclicity of the graph - a status we cannot continue working on -> if cyclic we get an error
538 |         #is_cyclic = SimplifyGraph.isCyclic(DG)
539 |         #if is_cyclic:
540 |         #    k_size+=1
541 |         #    w+=1
542 |         #    print("The graph has a cycle - critical error")
543 |             #return -1
544 |         #else:
545 |         #    print("No cycle in graph")
546 |         if DEBUG==True:
547 |             print("BATCHID",batch_id)
548 |             for id, value in all_batch_reads_dict.items():
549 |                 for other_id,other_val in value.items():
550 |                    print(id,": ",other_id,":",other_val[0],"::",other_val[1])
551 | 
552 |         mode = args.slow
553 |         #profiler = Profiler()
554 |         #profiler.start()
555 |         #the bubble popping step: We simplify the graph by linearizing all poppable bubbles
556 |         SimplifyGraph.simplifyGraph(DG, new_all_reads, work_dir, k_size, delta_len, mode)
557 |         #profiler.stop()
558 | 
559 |         #profiler.print()
560 |         #TODO: add delta as user parameter possibly?
561 |         delta = 0.15
562 |         print("Starting to generate Isoforms")
563 | 
564 |         if args.parallel:
565 |             batch_id = p_batch_id
566 |         #profiler = Profiler()
567 |         #profiler.start()
568 |         #generation of isoforms from the graph structure
569 |         IsoformGeneration.generate_isoforms(DG, new_all_reads, reads_for_isoforms, work_dir, outfolder, batch_id, delta, delta_len, delta_iso_len_3, delta_iso_len_5, max_seqs_to_spoa)
570 |         #profiler.stop()
571 | 
572 |         print("Isoforms generated-Starting batch merging ")
573 |     if not args.parallel:
574 |             print("Merging the batches with linear strategy")
575 |             #merges the predictions from different batches
576 |             #batch_merging_parallel.join_back_via_batch_merging(args.outfolder, delta, args.delta_len, args.delta_iso_len_3, args.delta_iso_len_5,
577 |             #                                                   args.max_seqs_to_spoa, args.iso_abundance)
578 | 
579 |     print("removing temporary workdir")
580 |     #sys.stdout.close()
581 |     shutil.rmtree(work_dir)
582 | 
583 | DEBUG=False
584 | 
585 | #TODO: add low_output (bool) as well as delta as user parameters and remove slow, merge_sub_isoforms_3, merge_sub_isoforms_5
586 | if __name__ == '__main__':
587 |     parser = argparse.ArgumentParser(description="De novo error correction of long-read transcriptome reads",
588 |                                      formatter_class=argparse.ArgumentDefaultsHelpFormatter)
589 |     parser.add_argument('--version', action='version', version='%(prog)s 0.3.9')
590 |     parser.add_argument('--fastq', type=str, default=False, help='Path to input fastq file with reads')
591 | 
592 |     parser.add_argument('--k', type=int, default=20, help='Kmer size')
593 |     parser.add_argument('--w', type=int, default=31, help='Window size')
594 |     parser.add_argument('--xmin', type=int, default=18, help='Upper interval length')
595 |     parser.add_argument('--xmax', type=int, default=80, help='Lower interval length')
596 |     parser.add_argument('--T', type=float, default=0.1, help='Minimum fraction keeping substitution')
597 |     parser.add_argument('--exact', action="store_true", help='Get exact solution for WIS for every read (recalculating weights for each read (much slower but slightly more accuracy,\
598 |                                                                  not to be used for clusters with over ~500 reads)')
599 |     parser.add_argument('--disable_numpy', action="store_true",
600 |                         help='Do not require numpy to be installed, but this version is about 1.5x slower than with numpy.')
601 |     parser.add_argument('--delta_len', type=int, default=5, help='Maximum length difference between two reads intervals for which they would still be merged')
602 |     parser.add_argument('--max_seqs_to_spoa', type=int, default=200, help='Maximum number of seqs to spoa')
603 |     parser.add_argument('--max_seqs', type=int, default=1000,
604 |                         help='Maximum number of seqs to correct at a time (in case of large clusters).')
605 |     parser.add_argument('--exact_instance_limit', type=int, default=0,
606 |                         help='Activates slower exact mode for instance smaller than this limit')
607 |     parser.add_argument('--set_w_dynamically', action="store_true",
608 |                         help='Set w = k + max(2*k, floor(cluster_size/1000)).')
609 |     parser.add_argument('--verbose', action="store_true", help='Print various developer stats.')
610 | 
611 |     parser.add_argument('--compression', action="store_true", help='Use homopolymer compressed reads. (Deprecated, because we will have fewer \
612 |                                                                         minmimizer combinations to span regions in homopolymenr dense regions. Solution \
613 |                                                                         could be to adjust upper interval length dynamically to guarantee a certain number of spanning intervals.')
614 |     parser.add_argument('--outfolder', type=str, default=None,
615 |                         help='The outfolder of isONform, into which the algorithm will write the results.')
616 |     parser.add_argument('--iso_abundance', type=int,default=5, help='Cutoff parameter: abundance of reads that have to support an isoform to show in results')
617 |     parser.add_argument('--delta_iso_len_3', type=int, default=30,
618 |                         help='Cutoff parameter: maximum length difference at 3prime end, for which subisoforms are still merged into longer isoforms')
619 |     parser.add_argument('--delta_iso_len_5', type=int, default=50,
620 |                         help='Cutoff parameter: maximum length difference at 5prime end, for which subisoforms are still merged into longer isoforms')
621 |     parser.add_argument('--parallel',type=bool,default=False,help='Indicates whether we run the parallelization wrapper script')
622 |     parser.add_argument('--slow',action="store_true", help='EXPERIMENTAL PARAMETER: has high repercussions on run time use the slow mode for the simplification of the graph (bubble popping), slow mode: every bubble gets popped.')
623 |     args = parser.parse_args()
624 | 
625 |     if args.xmin < 2 * args.k:
626 |         args.xmin = 2 * args.k
627 |         eprint("xmin set to {0}".format(args.xmin))
628 | 
629 |     if len(sys.argv) == 1:
630 |         parser.print_help()
631 |         sys.exit()
632 | 
633 |     os.environ['PYTHONHASHSEED'] = '0'
634 |     if args.outfolder and not os.path.exists(args.outfolder):
635 |         os.makedirs(args.outfolder)
636 | 
637 |     if 100 < args.w or args.w < args.k:
638 |         eprint('Please specify a window of size larger or equal to k, and smaller than 100.')
639 |         sys.exit(1)
640 | 
641 |     main(args)
642 | 


--------------------------------------------------------------------------------
/LICENSE:
--------------------------------------------------------------------------------
  1 |                     GNU GENERAL PUBLIC LICENSE
  2 |                        Version 3, 29 June 2007
  3 | 
  4 |  Copyright (C) 2007 Free Software Foundation, Inc. <https://fsf.org/>
  5 |  Everyone is permitted to copy and distribute verbatim copies
  6 |  of this license document, but changing it is not allowed.
  7 | 
  8 |                             Preamble
  9 | 
 10 |   The GNU General Public License is a free, copyleft license for
 11 | software and other kinds of works.
 12 | 
 13 |   The licenses for most software and other practical works are designed
 14 | to take away your freedom to share and change the works.  By contrast,
 15 | the GNU General Public License is intended to guarantee your freedom to
 16 | share and change all versions of a program--to make sure it remains free
 17 | software for all its users.  We, the Free Software Foundation, use the
 18 | GNU General Public License for most of our software; it applies also to
 19 | any other work released this way by its authors.  You can apply it to
 20 | your programs, too.
 21 | 
 22 |   When we speak of free software, we are referring to freedom, not
 23 | price.  Our General Public Licenses are designed to make sure that you
 24 | have the freedom to distribute copies of free software (and charge for
 25 | them if you wish), that you receive source code or can get it if you
 26 | want it, that you can change the software or use pieces of it in new
 27 | free programs, and that you know you can do these things.
 28 | 
 29 |   To protect your rights, we need to prevent others from denying you
 30 | these rights or asking you to surrender the rights.  Therefore, you have
 31 | certain responsibilities if you distribute copies of the software, or if
 32 | you modify it: responsibilities to respect the freedom of others.
 33 | 
 34 |   For example, if you distribute copies of such a program, whether
 35 | gratis or for a fee, you must pass on to the recipients the same
 36 | freedoms that you received.  You must make sure that they, too, receive
 37 | or can get the source code.  And you must show them these terms so they
 38 | know their rights.
 39 | 
 40 |   Developers that use the GNU GPL protect your rights with two steps:
 41 | (1) assert copyright on the software, and (2) offer you this License
 42 | giving you legal permission to copy, distribute and/or modify it.
 43 | 
 44 |   For the developers' and authors' protection, the GPL clearly explains
 45 | that there is no warranty for this free software.  For both users' and
 46 | authors' sake, the GPL requires that modified versions be marked as
 47 | changed, so that their problems will not be attributed erroneously to
 48 | authors of previous versions.
 49 | 
 50 |   Some devices are designed to deny users access to install or run
 51 | modified versions of the software inside them, although the manufacturer
 52 | can do so.  This is fundamentally incompatible with the aim of
 53 | protecting users' freedom to change the software.  The systematic
 54 | pattern of such abuse occurs in the area of products for individuals to
 55 | use, which is precisely where it is most unacceptable.  Therefore, we
 56 | have designed this version of the GPL to prohibit the practice for those
 57 | products.  If such problems arise substantially in other domains, we
 58 | stand ready to extend this provision to those domains in future versions
 59 | of the GPL, as needed to protect the freedom of users.
 60 | 
 61 |   Finally, every program is threatened constantly by software patents.
 62 | States should not allow patents to restrict development and use of
 63 | software on general-purpose computers, but in those that do, we wish to
 64 | avoid the special danger that patents applied to a free program could
 65 | make it effectively proprietary.  To prevent this, the GPL assures that
 66 | patents cannot be used to render the program non-free.
 67 | 
 68 |   The precise terms and conditions for copying, distribution and
 69 | modification follow.
 70 | 
 71 |                        TERMS AND CONDITIONS
 72 | 
 73 |   0. Definitions.
 74 | 
 75 |   "This License" refers to version 3 of the GNU General Public License.
 76 | 
 77 |   "Copyright" also means copyright-like laws that apply to other kinds of
 78 | works, such as semiconductor masks.
 79 | 
 80 |   "The Program" refers to any copyrightable work licensed under this
 81 | License.  Each licensee is addressed as "you".  "Licensees" and
 82 | "recipients" may be individuals or organizations.
 83 | 
 84 |   To "modify" a work means to copy from or adapt all or part of the work
 85 | in a fashion requiring copyright permission, other than the making of an
 86 | exact copy.  The resulting work is called a "modified version" of the
 87 | earlier work or a work "based on" the earlier work.
 88 | 
 89 |   A "covered work" means either the unmodified Program or a work based
 90 | on the Program.
 91 | 
 92 |   To "propagate" a work means to do anything with it that, without
 93 | permission, would make you directly or secondarily liable for
 94 | infringement under applicable copyright law, except executing it on a
 95 | computer or modifying a private copy.  Propagation includes copying,
 96 | distribution (with or without modification), making available to the
 97 | public, and in some countries other activities as well.
 98 | 
 99 |   To "convey" a work means any kind of propagation that enables other
100 | parties to make or receive copies.  Mere interaction with a user through
101 | a computer network, with no transfer of a copy, is not conveying.
102 | 
103 |   An interactive user interface displays "Appropriate Legal Notices"
104 | to the extent that it includes a convenient and prominently visible
105 | feature that (1) displays an appropriate copyright notice, and (2)
106 | tells the user that there is no warranty for the work (except to the
107 | extent that warranties are provided), that licensees may convey the
108 | work under this License, and how to view a copy of this License.  If
109 | the interface presents a list of user commands or options, such as a
110 | menu, a prominent item in the list meets this criterion.
111 | 
112 |   1. Source Code.
113 | 
114 |   The "source code" for a work means the preferred form of the work
115 | for making modifications to it.  "Object code" means any non-source
116 | form of a work.
117 | 
118 |   A "Standard Interface" means an interface that either is an official
119 | standard defined by a recognized standards body, or, in the case of
120 | interfaces specified for a particular programming language, one that
121 | is widely used among developers working in that language.
122 | 
123 |   The "System Libraries" of an executable work include anything, other
124 | than the work as a whole, that (a) is included in the normal form of
125 | packaging a Major Component, but which is not part of that Major
126 | Component, and (b) serves only to enable use of the work with that
127 | Major Component, or to implement a Standard Interface for which an
128 | implementation is available to the public in source code form.  A
129 | "Major Component", in this context, means a major essential component
130 | (kernel, window system, and so on) of the specific operating system
131 | (if any) on which the executable work runs, or a compiler used to
132 | produce the work, or an object code interpreter used to run it.
133 | 
134 |   The "Corresponding Source" for a work in object code form means all
135 | the source code needed to generate, install, and (for an executable
136 | work) run the object code and to modify the work, including scripts to
137 | control those activities.  However, it does not include the work's
138 | System Libraries, or general-purpose tools or generally available free
139 | programs which are used unmodified in performing those activities but
140 | which are not part of the work.  For example, Corresponding Source
141 | includes interface definition files associated with source files for
142 | the work, and the source code for shared libraries and dynamically
143 | linked subprograms that the work is specifically designed to require,
144 | such as by intimate data communication or control flow between those
145 | subprograms and other parts of the work.
146 | 
147 |   The Corresponding Source need not include anything that users
148 | can regenerate automatically from other parts of the Corresponding
149 | Source.
150 | 
151 |   The Corresponding Source for a work in source code form is that
152 | same work.
153 | 
154 |   2. Basic Permissions.
155 | 
156 |   All rights granted under this License are granted for the term of
157 | copyright on the Program, and are irrevocable provided the stated
158 | conditions are met.  This License explicitly affirms your unlimited
159 | permission to run the unmodified Program.  The output from running a
160 | covered work is covered by this License only if the output, given its
161 | content, constitutes a covered work.  This License acknowledges your
162 | rights of fair use or other equivalent, as provided by copyright law.
163 | 
164 |   You may make, run and propagate covered works that you do not
165 | convey, without conditions so long as your license otherwise remains
166 | in force.  You may convey covered works to others for the sole purpose
167 | of having them make modifications exclusively for you, or provide you
168 | with facilities for running those works, provided that you comply with
169 | the terms of this License in conveying all material for which you do
170 | not control copyright.  Those thus making or running the covered works
171 | for you must do so exclusively on your behalf, under your direction
172 | and control, on terms that prohibit them from making any copies of
173 | your copyrighted material outside their relationship with you.
174 | 
175 |   Conveying under any other circumstances is permitted solely under
176 | the conditions stated below.  Sublicensing is not allowed; section 10
177 | makes it unnecessary.
178 | 
179 |   3. Protecting Users' Legal Rights From Anti-Circumvention Law.
180 | 
181 |   No covered work shall be deemed part of an effective technological
182 | measure under any applicable law fulfilling obligations under article
183 | 11 of the WIPO copyright treaty adopted on 20 December 1996, or
184 | similar laws prohibiting or restricting circumvention of such
185 | measures.
186 | 
187 |   When you convey a covered work, you waive any legal power to forbid
188 | circumvention of technological measures to the extent such circumvention
189 | is effected by exercising rights under this License with respect to
190 | the covered work, and you disclaim any intention to limit operation or
191 | modification of the work as a means of enforcing, against the work's
192 | users, your or third parties' legal rights to forbid circumvention of
193 | technological measures.
194 | 
195 |   4. Conveying Verbatim Copies.
196 | 
197 |   You may convey verbatim copies of the Program's source code as you
198 | receive it, in any medium, provided that you conspicuously and
199 | appropriately publish on each copy an appropriate copyright notice;
200 | keep intact all notices stating that this License and any
201 | non-permissive terms added in accord with section 7 apply to the code;
202 | keep intact all notices of the absence of any warranty; and give all
203 | recipients a copy of this License along with the Program.
204 | 
205 |   You may charge any price or no price for each copy that you convey,
206 | and you may offer support or warranty protection for a fee.
207 | 
208 |   5. Conveying Modified Source Versions.
209 | 
210 |   You may convey a work based on the Program, or the modifications to
211 | produce it from the Program, in the form of source code under the
212 | terms of section 4, provided that you also meet all of these conditions:
213 | 
214 |     a) The work must carry prominent notices stating that you modified
215 |     it, and giving a relevant date.
216 | 
217 |     b) The work must carry prominent notices stating that it is
218 |     released under this License and any conditions added under section
219 |     7.  This requirement modifies the requirement in section 4 to
220 |     "keep intact all notices".
221 | 
222 |     c) You must license the entire work, as a whole, under this
223 |     License to anyone who comes into possession of a copy.  This
224 |     License will therefore apply, along with any applicable section 7
225 |     additional terms, to the whole of the work, and all its parts,
226 |     regardless of how they are packaged.  This License gives no
227 |     permission to license the work in any other way, but it does not
228 |     invalidate such permission if you have separately received it.
229 | 
230 |     d) If the work has interactive user interfaces, each must display
231 |     Appropriate Legal Notices; however, if the Program has interactive
232 |     interfaces that do not display Appropriate Legal Notices, your
233 |     work need not make them do so.
234 | 
235 |   A compilation of a covered work with other separate and independent
236 | works, which are not by their nature extensions of the covered work,
237 | and which are not combined with it such as to form a larger program,
238 | in or on a volume of a storage or distribution medium, is called an
239 | "aggregate" if the compilation and its resulting copyright are not
240 | used to limit the access or legal rights of the compilation's users
241 | beyond what the individual works permit.  Inclusion of a covered work
242 | in an aggregate does not cause this License to apply to the other
243 | parts of the aggregate.
244 | 
245 |   6. Conveying Non-Source Forms.
246 | 
247 |   You may convey a covered work in object code form under the terms
248 | of sections 4 and 5, provided that you also convey the
249 | machine-readable Corresponding Source under the terms of this License,
250 | in one of these ways:
251 | 
252 |     a) Convey the object code in, or embodied in, a physical product
253 |     (including a physical distribution medium), accompanied by the
254 |     Corresponding Source fixed on a durable physical medium
255 |     customarily used for software interchange.
256 | 
257 |     b) Convey the object code in, or embodied in, a physical product
258 |     (including a physical distribution medium), accompanied by a
259 |     written offer, valid for at least three years and valid for as
260 |     long as you offer spare parts or customer support for that product
261 |     model, to give anyone who possesses the object code either (1) a
262 |     copy of the Corresponding Source for all the software in the
263 |     product that is covered by this License, on a durable physical
264 |     medium customarily used for software interchange, for a price no
265 |     more than your reasonable cost of physically performing this
266 |     conveying of source, or (2) access to copy the
267 |     Corresponding Source from a network server at no charge.
268 | 
269 |     c) Convey individual copies of the object code with a copy of the
270 |     written offer to provide the Corresponding Source.  This
271 |     alternative is allowed only occasionally and noncommercially, and
272 |     only if you received the object code with such an offer, in accord
273 |     with subsection 6b.
274 | 
275 |     d) Convey the object code by offering access from a designated
276 |     place (gratis or for a charge), and offer equivalent access to the
277 |     Corresponding Source in the same way through the same place at no
278 |     further charge.  You need not require recipients to copy the
279 |     Corresponding Source along with the object code.  If the place to
280 |     copy the object code is a network server, the Corresponding Source
281 |     may be on a different server (operated by you or a third party)
282 |     that supports equivalent copying facilities, provided you maintain
283 |     clear directions next to the object code saying where to find the
284 |     Corresponding Source.  Regardless of what server hosts the
285 |     Corresponding Source, you remain obligated to ensure that it is
286 |     available for as long as needed to satisfy these requirements.
287 | 
288 |     e) Convey the object code using peer-to-peer transmission, provided
289 |     you inform other peers where the object code and Corresponding
290 |     Source of the work are being offered to the general public at no
291 |     charge under subsection 6d.
292 | 
293 |   A separable portion of the object code, whose source code is excluded
294 | from the Corresponding Source as a System Library, need not be
295 | included in conveying the object code work.
296 | 
297 |   A "User Product" is either (1) a "consumer product", which means any
298 | tangible personal property which is normally used for personal, family,
299 | or household purposes, or (2) anything designed or sold for incorporation
300 | into a dwelling.  In determining whether a product is a consumer product,
301 | doubtful cases shall be resolved in favor of coverage.  For a particular
302 | product received by a particular user, "normally used" refers to a
303 | typical or common use of that class of product, regardless of the status
304 | of the particular user or of the way in which the particular user
305 | actually uses, or expects or is expected to use, the product.  A product
306 | is a consumer product regardless of whether the product has substantial
307 | commercial, industrial or non-consumer uses, unless such uses represent
308 | the only significant mode of use of the product.
309 | 
310 |   "Installation Information" for a User Product means any methods,
311 | procedures, authorization keys, or other information required to install
312 | and execute modified versions of a covered work in that User Product from
313 | a modified version of its Corresponding Source.  The information must
314 | suffice to ensure that the continued functioning of the modified object
315 | code is in no case prevented or interfered with solely because
316 | modification has been made.
317 | 
318 |   If you convey an object code work under this section in, or with, or
319 | specifically for use in, a User Product, and the conveying occurs as
320 | part of a transaction in which the right of possession and use of the
321 | User Product is transferred to the recipient in perpetuity or for a
322 | fixed term (regardless of how the transaction is characterized), the
323 | Corresponding Source conveyed under this section must be accompanied
324 | by the Installation Information.  But this requirement does not apply
325 | if neither you nor any third party retains the ability to install
326 | modified object code on the User Product (for example, the work has
327 | been installed in ROM).
328 | 
329 |   The requirement to provide Installation Information does not include a
330 | requirement to continue to provide support service, warranty, or updates
331 | for a work that has been modified or installed by the recipient, or for
332 | the User Product in which it has been modified or installed.  Access to a
333 | network may be denied when the modification itself materially and
334 | adversely affects the operation of the network or violates the rules and
335 | protocols for communication across the network.
336 | 
337 |   Corresponding Source conveyed, and Installation Information provided,
338 | in accord with this section must be in a format that is publicly
339 | documented (and with an implementation available to the public in
340 | source code form), and must require no special password or key for
341 | unpacking, reading or copying.
342 | 
343 |   7. Additional Terms.
344 | 
345 |   "Additional permissions" are terms that supplement the terms of this
346 | License by making exceptions from one or more of its conditions.
347 | Additional permissions that are applicable to the entire Program shall
348 | be treated as though they were included in this License, to the extent
349 | that they are valid under applicable law.  If additional permissions
350 | apply only to part of the Program, that part may be used separately
351 | under those permissions, but the entire Program remains governed by
352 | this License without regard to the additional permissions.
353 | 
354 |   When you convey a copy of a covered work, you may at your option
355 | remove any additional permissions from that copy, or from any part of
356 | it.  (Additional permissions may be written to require their own
357 | removal in certain cases when you modify the work.)  You may place
358 | additional permissions on material, added by you to a covered work,
359 | for which you have or can give appropriate copyright permission.
360 | 
361 |   Notwithstanding any other provision of this License, for material you
362 | add to a covered work, you may (if authorized by the copyright holders of
363 | that material) supplement the terms of this License with terms:
364 | 
365 |     a) Disclaiming warranty or limiting liability differently from the
366 |     terms of sections 15 and 16 of this License; or
367 | 
368 |     b) Requiring preservation of specified reasonable legal notices or
369 |     author attributions in that material or in the Appropriate Legal
370 |     Notices displayed by works containing it; or
371 | 
372 |     c) Prohibiting misrepresentation of the origin of that material, or
373 |     requiring that modified versions of such material be marked in
374 |     reasonable ways as different from the original version; or
375 | 
376 |     d) Limiting the use for publicity purposes of names of licensors or
377 |     authors of the material; or
378 | 
379 |     e) Declining to grant rights under trademark law for use of some
380 |     trade names, trademarks, or service marks; or
381 | 
382 |     f) Requiring indemnification of licensors and authors of that
383 |     material by anyone who conveys the material (or modified versions of
384 |     it) with contractual assumptions of liability to the recipient, for
385 |     any liability that these contractual assumptions directly impose on
386 |     those licensors and authors.
387 | 
388 |   All other non-permissive additional terms are considered "further
389 | restrictions" within the meaning of section 10.  If the Program as you
390 | received it, or any part of it, contains a notice stating that it is
391 | governed by this License along with a term that is a further
392 | restriction, you may remove that term.  If a license document contains
393 | a further restriction but permits relicensing or conveying under this
394 | License, you may add to a covered work material governed by the terms
395 | of that license document, provided that the further restriction does
396 | not survive such relicensing or conveying.
397 | 
398 |   If you add terms to a covered work in accord with this section, you
399 | must place, in the relevant source files, a statement of the
400 | additional terms that apply to those files, or a notice indicating
401 | where to find the applicable terms.
402 | 
403 |   Additional terms, permissive or non-permissive, may be stated in the
404 | form of a separately written license, or stated as exceptions;
405 | the above requirements apply either way.
406 | 
407 |   8. Termination.
408 | 
409 |   You may not propagate or modify a covered work except as expressly
410 | provided under this License.  Any attempt otherwise to propagate or
411 | modify it is void, and will automatically terminate your rights under
412 | this License (including any patent licenses granted under the third
413 | paragraph of section 11).
414 | 
415 |   However, if you cease all violation of this License, then your
416 | license from a particular copyright holder is reinstated (a)
417 | provisionally, unless and until the copyright holder explicitly and
418 | finally terminates your license, and (b) permanently, if the copyright
419 | holder fails to notify you of the violation by some reasonable means
420 | prior to 60 days after the cessation.
421 | 
422 |   Moreover, your license from a particular copyright holder is
423 | reinstated permanently if the copyright holder notifies you of the
424 | violation by some reasonable means, this is the first time you have
425 | received notice of violation of this License (for any work) from that
426 | copyright holder, and you cure the violation prior to 30 days after
427 | your receipt of the notice.
428 | 
429 |   Termination of your rights under this section does not terminate the
430 | licenses of parties who have received copies or rights from you under
431 | this License.  If your rights have been terminated and not permanently
432 | reinstated, you do not qualify to receive new licenses for the same
433 | material under section 10.
434 | 
435 |   9. Acceptance Not Required for Having Copies.
436 | 
437 |   You are not required to accept this License in order to receive or
438 | run a copy of the Program.  Ancillary propagation of a covered work
439 | occurring solely as a consequence of using peer-to-peer transmission
440 | to receive a copy likewise does not require acceptance.  However,
441 | nothing other than this License grants you permission to propagate or
442 | modify any covered work.  These actions infringe copyright if you do
443 | not accept this License.  Therefore, by modifying or propagating a
444 | covered work, you indicate your acceptance of this License to do so.
445 | 
446 |   10. Automatic Licensing of Downstream Recipients.
447 | 
448 |   Each time you convey a covered work, the recipient automatically
449 | receives a license from the original licensors, to run, modify and
450 | propagate that work, subject to this License.  You are not responsible
451 | for enforcing compliance by third parties with this License.
452 | 
453 |   An "entity transaction" is a transaction transferring control of an
454 | organization, or substantially all assets of one, or subdividing an
455 | organization, or merging organizations.  If propagation of a covered
456 | work results from an entity transaction, each party to that
457 | transaction who receives a copy of the work also receives whatever
458 | licenses to the work the party's predecessor in interest had or could
459 | give under the previous paragraph, plus a right to possession of the
460 | Corresponding Source of the work from the predecessor in interest, if
461 | the predecessor has it or can get it with reasonable efforts.
462 | 
463 |   You may not impose any further restrictions on the exercise of the
464 | rights granted or affirmed under this License.  For example, you may
465 | not impose a license fee, royalty, or other charge for exercise of
466 | rights granted under this License, and you may not initiate litigation
467 | (including a cross-claim or counterclaim in a lawsuit) alleging that
468 | any patent claim is infringed by making, using, selling, offering for
469 | sale, or importing the Program or any portion of it.
470 | 
471 |   11. Patents.
472 | 
473 |   A "contributor" is a copyright holder who authorizes use under this
474 | License of the Program or a work on which the Program is based.  The
475 | work thus licensed is called the contributor's "contributor version".
476 | 
477 |   A contributor's "essential patent claims" are all patent claims
478 | owned or controlled by the contributor, whether already acquired or
479 | hereafter acquired, that would be infringed by some manner, permitted
480 | by this License, of making, using, or selling its contributor version,
481 | but do not include claims that would be infringed only as a
482 | consequence of further modification of the contributor version.  For
483 | purposes of this definition, "control" includes the right to grant
484 | patent sublicenses in a manner consistent with the requirements of
485 | this License.
486 | 
487 |   Each contributor grants you a non-exclusive, worldwide, royalty-free
488 | patent license under the contributor's essential patent claims, to
489 | make, use, sell, offer for sale, import and otherwise run, modify and
490 | propagate the contents of its contributor version.
491 | 
492 |   In the following three paragraphs, a "patent license" is any express
493 | agreement or commitment, however denominated, not to enforce a patent
494 | (such as an express permission to practice a patent or covenant not to
495 | sue for patent infringement).  To "grant" such a patent license to a
496 | party means to make such an agreement or commitment not to enforce a
497 | patent against the party.
498 | 
499 |   If you convey a covered work, knowingly relying on a patent license,
500 | and the Corresponding Source of the work is not available for anyone
501 | to copy, free of charge and under the terms of this License, through a
502 | publicly available network server or other readily accessible means,
503 | then you must either (1) cause the Corresponding Source to be so
504 | available, or (2) arrange to deprive yourself of the benefit of the
505 | patent license for this particular work, or (3) arrange, in a manner
506 | consistent with the requirements of this License, to extend the patent
507 | license to downstream recipients.  "Knowingly relying" means you have
508 | actual knowledge that, but for the patent license, your conveying the
509 | covered work in a country, or your recipient's use of the covered work
510 | in a country, would infringe one or more identifiable patents in that
511 | country that you have reason to believe are valid.
512 | 
513 |   If, pursuant to or in connection with a single transaction or
514 | arrangement, you convey, or propagate by procuring conveyance of, a
515 | covered work, and grant a patent license to some of the parties
516 | receiving the covered work authorizing them to use, propagate, modify
517 | or convey a specific copy of the covered work, then the patent license
518 | you grant is automatically extended to all recipients of the covered
519 | work and works based on it.
520 | 
521 |   A patent license is "discriminatory" if it does not include within
522 | the scope of its coverage, prohibits the exercise of, or is
523 | conditioned on the non-exercise of one or more of the rights that are
524 | specifically granted under this License.  You may not convey a covered
525 | work if you are a party to an arrangement with a third party that is
526 | in the business of distributing software, under which you make payment
527 | to the third party based on the extent of your activity of conveying
528 | the work, and under which the third party grants, to any of the
529 | parties who would receive the covered work from you, a discriminatory
530 | patent license (a) in connection with copies of the covered work
531 | conveyed by you (or copies made from those copies), or (b) primarily
532 | for and in connection with specific products or compilations that
533 | contain the covered work, unless you entered into that arrangement,
534 | or that patent license was granted, prior to 28 March 2007.
535 | 
536 |   Nothing in this License shall be construed as excluding or limiting
537 | any implied license or other defenses to infringement that may
538 | otherwise be available to you under applicable patent law.
539 | 
540 |   12. No Surrender of Others' Freedom.
541 | 
542 |   If conditions are imposed on you (whether by court order, agreement or
543 | otherwise) that contradict the conditions of this License, they do not
544 | excuse you from the conditions of this License.  If you cannot convey a
545 | covered work so as to satisfy simultaneously your obligations under this
546 | License and any other pertinent obligations, then as a consequence you may
547 | not convey it at all.  For example, if you agree to terms that obligate you
548 | to collect a royalty for further conveying from those to whom you convey
549 | the Program, the only way you could satisfy both those terms and this
550 | License would be to refrain entirely from conveying the Program.
551 | 
552 |   13. Use with the GNU Affero General Public License.
553 | 
554 |   Notwithstanding any other provision of this License, you have
555 | permission to link or combine any covered work with a work licensed
556 | under version 3 of the GNU Affero General Public License into a single
557 | combined work, and to convey the resulting work.  The terms of this
558 | License will continue to apply to the part which is the covered work,
559 | but the special requirements of the GNU Affero General Public License,
560 | section 13, concerning interaction through a network will apply to the
561 | combination as such.
562 | 
563 |   14. Revised Versions of this License.
564 | 
565 |   The Free Software Foundation may publish revised and/or new versions of
566 | the GNU General Public License from time to time.  Such new versions will
567 | be similar in spirit to the present version, but may differ in detail to
568 | address new problems or concerns.
569 | 
570 |   Each version is given a distinguishing version number.  If the
571 | Program specifies that a certain numbered version of the GNU General
572 | Public License "or any later version" applies to it, you have the
573 | option of following the terms and conditions either of that numbered
574 | version or of any later version published by the Free Software
575 | Foundation.  If the Program does not specify a version number of the
576 | GNU General Public License, you may choose any version ever published
577 | by the Free Software Foundation.
578 | 
579 |   If the Program specifies that a proxy can decide which future
580 | versions of the GNU General Public License can be used, that proxy's
581 | public statement of acceptance of a version permanently authorizes you
582 | to choose that version for the Program.
583 | 
584 |   Later license versions may give you additional or different
585 | permissions.  However, no additional obligations are imposed on any
586 | author or copyright holder as a result of your choosing to follow a
587 | later version.
588 | 
589 |   15. Disclaimer of Warranty.
590 | 
591 |   THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY
592 | APPLICABLE LAW.  EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT
593 | HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY
594 | OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO,
595 | THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
596 | PURPOSE.  THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM
597 | IS WITH YOU.  SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF
598 | ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
599 | 
600 |   16. Limitation of Liability.
601 | 
602 |   IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
603 | WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS
604 | THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY
605 | GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE
606 | USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF
607 | DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD
608 | PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS),
609 | EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF
610 | SUCH DAMAGES.
611 | 
612 |   17. Interpretation of Sections 15 and 16.
613 | 
614 |   If the disclaimer of warranty and limitation of liability provided
615 | above cannot be given local legal effect according to their terms,
616 | reviewing courts shall apply local law that most closely approximates
617 | an absolute waiver of all civil liability in connection with the
618 | Program, unless a warranty or assumption of liability accompanies a
619 | copy of the Program in return for a fee.
620 | 
621 |                      END OF TERMS AND CONDITIONS
622 | 
623 |             How to Apply These Terms to Your New Programs
624 | 
625 |   If you develop a new program, and you want it to be of the greatest
626 | possible use to the public, the best way to achieve this is to make it
627 | free software which everyone can redistribute and change under these terms.
628 | 
629 |   To do so, attach the following notices to the program.  It is safest
630 | to attach them to the start of each source file to most effectively
631 | state the exclusion of warranty; and each file should have at least
632 | the "copyright" line and a pointer to where the full notice is found.
633 | 
634 |     <one line to give the program's name and a brief idea of what it does.>
635 |     Copyright (C) <year>  <name of author>
636 | 
637 |     This program is free software: you can redistribute it and/or modify
638 |     it under the terms of the GNU General Public License as published by
639 |     the Free Software Foundation, either version 3 of the License, or
640 |     (at your option) any later version.
641 | 
642 |     This program is distributed in the hope that it will be useful,
643 |     but WITHOUT ANY WARRANTY; without even the implied warranty of
644 |     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
645 |     GNU General Public License for more details.
646 | 
647 |     You should have received a copy of the GNU General Public License
648 |     along with this program.  If not, see <https://www.gnu.org/licenses/>.
649 | 
650 | Also add information on how to contact you by electronic and paper mail.
651 | 
652 |   If the program does terminal interaction, make it output a short
653 | notice like this when it starts in an interactive mode:
654 | 
655 |     <program>  Copyright (C) <year>  <name of author>
656 |     This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
657 |     This is free software, and you are welcome to redistribute it
658 |     under certain conditions; type `show c' for details.
659 | 
660 | The hypothetical commands `show w' and `show c' should show the appropriate
661 | parts of the General Public License.  Of course, your program's commands
662 | might be different; for a GUI interface, you would use an "about box".
663 | 
664 |   You should also get your employer (if you work as a programmer) or school,
665 | if any, to sign a "copyright disclaimer" for the program, if necessary.
666 | For more information on this, and how to apply and follow the GNU GPL, see
667 | <https://www.gnu.org/licenses/>.
668 | 
669 |   The GNU General Public License does not permit incorporating your program
670 | into proprietary programs.  If your program is a subroutine library, you
671 | may consider it more useful to permit linking proprietary applications with
672 | the library.  If this is what you want to do, use the GNU Lesser General
673 | Public License instead of this License.  But first, please read
674 | <https://www.gnu.org/licenses/why-not-lgpl.html>.
675 | 


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