├── .gitignore ├── LICENSE ├── README.md ├── feature_importances ├── SPEID │ └── from_HOCOMOCO_motifs │ │ ├── GM12878_enhancers_feature_importance.csv │ │ ├── GM12878_promoters_feature_importance.csv │ │ ├── HUVEC_enhancers_feature_importance.csv │ │ ├── HUVEC_promoters_feature_importance.csv │ │ ├── HeLa-S3_enhancers_feature_importance.csv │ │ ├── HeLa-S3_promoters_feature_importance.csv │ │ ├── IMR90_enhancers_feature_importance.csv │ │ ├── IMR90_promoters_feature_importance.csv │ │ ├── K562_enhancers_feature_importance.csv │ │ ├── K562_promoters_feature_importance.csv │ │ ├── NHEK_enhancers_feature_importance.csv │ │ └── NHEK_promoters_feature_importance.csv ├── analyze_feature_importances.m ├── collect_SPEID_results.m ├── name_scatter_plot.m └── read_SPEID_feature_importance.m ├── figs ├── SPEID_versus_TargetFinder.pdf ├── importance_all.pdf ├── importance_hist.pdf ├── importance_over_count.pdf ├── importance_top_20.pdf ├── mean_diff_all.pdf ├── mean_diff_hist.pdf ├── mean_diff_over_count.pdf └── mean_diff_top_20.pdf └── pairwise ├── basic_training.py ├── build_small_model.py ├── data_processing ├── combine_hdf5s.py ├── load_data_pairs.py ├── txt_to_hdf5_ep_split.py ├── write_predictions_to_CSV.py └── write_sequences_to_fasta.py ├── fimo_all_GM12878_motifs.sh ├── frozen_model ├── build_model.py └── retraining_model3.py ├── load_data_pairs.py ├── predict.py ├── read_FIMO_results.py └── util.py /.gitignore: -------------------------------------------------------------------------------- 1 | # Old stuff 2 | old/* 3 | feature_importances/SPEID/old/* 4 | pairwise/old_models/* 5 | cluster_installation_steps.sh 6 | pairwise/FIMO 7 | pairwise/motifs 8 | build_smaller_model.py 9 | 10 | # Dependencies 11 | dependencies/ 12 | 13 | # Data files 14 | data/* 15 | *.mat 16 | *.zip 17 | 18 | # Network Weight Files 19 | *.hdf5 20 | 21 | # Byte-compiled / optimized / DLL files 22 | __pycache__/ 23 | *.py[cod] 24 | *$py.class 25 | 26 | # C extensions 27 | *.so 28 | 29 | # Distribution / packaging 30 | .Python 31 | env/ 32 | build/ 33 | develop-eggs/ 34 | dist/ 35 | downloads/ 36 | eggs/ 37 | .eggs/ 38 | lib/ 39 | lib64/ 40 | parts/ 41 | sdist/ 42 | var/ 43 | *.egg-info/ 44 | .installed.cfg 45 | *.egg 46 | 47 | # PyInstaller 48 | # Usually these files are written by a python script from a template 49 | # before PyInstaller builds the exe, so as to inject date/other infos into it. 50 | *.manifest 51 | *.spec 52 | 53 | # Installer logs 54 | pip-log.txt 55 | pip-delete-this-directory.txt 56 | 57 | # Unit test / coverage reports 58 | htmlcov/ 59 | .tox/ 60 | .coverage 61 | .coverage.* 62 | .cache 63 | nosetests.xml 64 | coverage.xml 65 | *,cover 66 | .hypothesis/ 67 | 68 | # Translations 69 | *.mo 70 | *.pot 71 | 72 | # Django stuff: 73 | *.log 74 | 75 | # Sphinx documentation 76 | docs/_build/ 77 | 78 | # PyBuilder 79 | target/ 80 | 81 | #Ipython Notebook 82 | .ipynb_checkpoints 83 | -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | GNU GENERAL PUBLIC LICENSE 2 | Version 3, 29 June 2007 3 | 4 | Copyright (C) 2007 Free Software Foundation, Inc. 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 . 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 | {project} Copyright (C) {year} {fullname} 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 | . 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 | . 675 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # SPEID 2 | 3 |

Overview

4 | 5 | SPEID is a deep neural network (implemented in Python, based on the Theano and Keras libraries) designed for predicting enhancer-promoter interactions (EPIs) directly from sequence data, as described in 6 | 7 | Singh, Shashank, Yang, Yang, Poczos, Barnabas, and Ma, Jian. "Predicting Enhancer-Promoter Interaction from Genomic Sequence with Deep Neural Networks." bioRxiv (2016): 085241. http://biorxiv.org/content/early/2016/11/02/085241. 8 | 9 |

Data Requirements

10 | The main data requirement for training SPEID is DNA sequences of the positive and negative enhancer promoter pairs in the desired cell type. For convenient reproducability in the cell lines we studied (GM12878, HeLa-S3, HUVEC, IMR90, K562, and NHEK) we have packaged these data together in a single HDF5 file available at http://genome.compbio.cs.cmu.edu/~sss1/SPEID/all_sequence_data.h5 (note that this is a large-ish (31GB) file). 11 | 12 | One can optionally initialize a number of the convolution kernels with motifs from the JASPAR database. These motifs can be conveniently accessed from the numpy file available at https://github.com/uci-cbcl/DanQ/blob/master/JASPAR_CORE_2016_vertebrates.npy. 13 | 14 |

SPEID Training/Prediction

15 | 16 | Running basic_training.py will train a model for each cell line and save the learned weights. Note that training SPEID is prohibitively slow without a GPU, and, even with a GPU, can take between a few hours and a few days (per cell line), depending on the GPU. On our setup (using a single NVIDIA GTX 1080), SPEID takes about 20 hours (per cell line) to train. 17 | 18 | If you are only interested in using SPEID for prediction, you can download our learned weights from http://genome.compbio.cs.cmu.edu/~sss1/SPEID/ and run [TODO: Add simple file for runnning predictions.]. Note that prediction is much faster than training, taking a few minutes with a decent GPU, and a few hours otherwise. 19 | 20 |

Estimating Feature Importance with SPEID

21 | 22 | By predicting interactions between arbitrary enhancer and promoter sequences, SPEID allows us to measure the effects of any particular sequence feature on EPI prediction, providing a general method for evaluating the effects of any sequence modification. We specifically applied this to measure importance of human sequence motifs from the HOCOMOCOv10 database (http://hocomoco.autosome.ru/), by measuring the change in prediction performance when all occurences of that feature are removed (i.e., replaced with noise). Our implementation depended on the FIMO motif scanning algorithm, included as part of the MEME-Suite collection (http://meme-suite.org/tools/fimo). The pairwise/FIMO directory contains code required for estimating feature importances, given a set of feature locations. Note that this step can be computationally intensive, as it requries running the model prediction on all test data points (for each sequence feature, cell line, etc.). The feature_importances directory contains MATLAB code for then analyzing the results. 23 | -------------------------------------------------------------------------------- /feature_importances/SPEID/from_HOCOMOCO_motifs/K562_enhancers_feature_importance.csv: -------------------------------------------------------------------------------- 1 | Motif Name,Motif Count,AUPR Difference,MS Difference 2 | ETS1_HUMAN.H10MO.C,17177,0.00054351456114,-3.59714e-05 3 | ATF2_HUMAN.H10MO.B,21805,0.000469201347327,1.15931e-05 4 | FOXD3_HUMAN.H10MO.D,56920,0.00127773237466,0.000322253 5 | ZBTB6_HUMAN.H10MO.D,16784,0.000423510295185,1.04606e-05 6 | NR1H2_HUMAN.H10MO.D,19202,0.000376536335502,7.56979e-06 7 | ETV2_HUMAN.H10MO.D,46943,0.00111229215793,0.000158519 8 | ETV3_HUMAN.H10MO.D,30877,0.000746918879038,-1.40369e-05 9 | RFX5_HUMAN.H10MO.A,27586,0.00035631255275,3.30508e-05 10 | SCRT2_HUMAN.H10MO.D,26008,0.000504189300907,-1.25766e-05 11 | ELF2_HUMAN.H10MO.C,70008,0.0014556852364,0.000242203 12 | NRF1_HUMAN.H10MO.A,83329,0.00133176673365,0.000252038 13 | MLXPL_HUMAN.H10MO.D,47650,0.00123680196523,-6.83665e-05 14 | PAX6_HUMAN.H10MO.D,24476,0.000651949748509,-1.93417e-05 15 | MAX_HUMAN.H10MO.A,51803,0.00117276934953,-2.44975e-05 16 | KLF6_HUMAN.H10MO.D,78110,0.00172654376221,0.00024426 17 | HNF1B_HUMAN.H10MO.B,23777,0.000782613064199,2.3216e-05 18 | PBX2_HUMAN.H10MO.C,19165,0.000541815745376,4.16636e-05 19 | RXRA_HUMAN.H10MO.C,42672,0.00093790469809,6.04093e-05 20 | GLIS2_HUMAN.H10MO.D,67142,0.00142777934387,0.000479609 21 | NFKB2_HUMAN.H10MO.D,22475,0.000619684146255,-8.43406e-06 22 | HSF4_HUMAN.H10MO.D,7938,0.000141421642723,-2.5183e-05 23 | ERR2_HUMAN.H10MO.A,11295,0.000396584205808,-1.62423e-05 24 | ESX1_HUMAN.H10MO.D,33130,0.000646005294604,6.65784e-05 25 | AP2B_HUMAN.H10MO.B,92917,0.00197504537535,0.000159502 26 | SPI1_HUMAN.H10MO.A,66957,0.00169795964771,4.42564e-05 27 | MYBA_HUMAN.H10MO.D,36367,0.000667388730611,-9.74536e-06 28 | GFI1B_HUMAN.H10MO.C,14635,0.00023775231107,1.48714e-05 29 | PO6F1_HUMAN.H10MO.D,29095,0.000685630243615,7.86781e-06 30 | ERR1_HUMAN.H10MO.D,59638,0.00147163920408,4.04119e-05 31 | PLAG1_HUMAN.H10MO.S,63111,0.00187099247513,0.000245303 32 | MAFA_HUMAN.H10MO.D,20450,0.00049325807857,-2.74181e-06 33 | ELF1_HUMAN.H10MO.A,45494,0.000965370861417,0.00012219 34 | SPDEF_HUMAN.H10MO.D,24988,0.000706522897978,-8.34465e-06 35 | GRHL1_HUMAN.H10MO.D,27298,0.000544494392532,2.50936e-05 36 | SCRT1_HUMAN.H10MO.D,19916,0.000609557682035,-4.29153e-06 37 | FOXJ2_HUMAN.H10MO.C,57794,0.00119012025709,0.000274539 38 | ASCL2_HUMAN.H10MO.D,61020,0.00154101926117,0.00035879 39 | XBP1_HUMAN.H10MO.C,45619,0.00107143702174,1.01328e-05 40 | FOXF2_HUMAN.H10MO.D,31343,0.000640330106514,0.000110149 41 | NFIC_HUMAN.H10MO.A,23335,0.000424260940958,-3.27826e-07 42 | MYC_HUMAN.H10MO.A,63401,0.00139437114829,-3.8147e-06 43 | STA5B_HUMAN.H10MO.C,7928,0.000238022166119,2.17557e-06 44 | RELB_HUMAN.H10MO.C,25396,0.000831480638524,-2.79546e-05 45 | HNF4G_HUMAN.H10MO.C,26296,0.000484944400713,4.38094e-06 46 | IRF4_HUMAN.H10MO.C,72842,0.001828873166,0.000127286 47 | MEOX2_HUMAN.H10MO.D,29252,0.000915320619418,3.85642e-05 48 | DBP_HUMAN.H10MO.B,21861,0.000399089745322,1.08778e-05 49 | ANDR_HUMAN.H10MO.A,36671,0.000730139619135,8.39531e-05 50 | P63_HUMAN.H10MO.A,19560,0.000390237192824,-8.04663e-07 51 | AIRE_HUMAN.H10MO.C,29035,0.00050442420955,5.56111e-05 52 | NFYA_HUMAN.H10MO.A,48938,0.00101001404632,9.60231e-05 53 | PO5F1_HUMAN.H10MO.A,31737,0.000903119813782,5.54621e-05 54 | SOX8_HUMAN.H10MO.D,17982,0.00056976140237,3.59714e-05 55 | THA_HUMAN.H10MO.C,21520,0.000514003710306,8.9407e-07 56 | INSM1_HUMAN.H10MO.C,73335,0.00151724290887,0.0001553 57 | CDX2_HUMAN.H10MO.C,21535,0.000558419843287,2.94745e-05 58 | ETS2_HUMAN.H10MO.C,56949,0.00124947325111,0.000123709 59 | ZN713_HUMAN.H10MO.D,51616,0.00132335077829,0.000264794 60 | MYF6_HUMAN.H10MO.C,13985,0.000396511365579,-2.05636e-05 61 | RX_HUMAN.H10MO.D,28312,0.000523716365062,5.47767e-05 62 | GATA3_HUMAN.H10MO.C,19244,0.000651624479069,4.82798e-05 63 | ZN148_HUMAN.H10MO.D,94941,0.000951059785948,0.000377208 64 | PO3F4_HUMAN.H10MO.D,32871,0.000834696331659,5.65648e-05 65 | ZN784_HUMAN.H10MO.D,33542,0.000831384250204,9.20594e-05 66 | ZN524_HUMAN.H10MO.D,74646,0.00154011631095,0.000231832 67 | STAT6_HUMAN.H10MO.C,11175,0.000335639679498,-7.7188e-06 68 | PAX1_HUMAN.H10MO.D,29125,0.000812825477792,-1.19209e-05 69 | PRRX1_HUMAN.H10MO.D,6027,0.000104439718552,-4.85778e-06 70 | ZIC4_HUMAN.H10MO.D,85648,0.00190636186684,0.000197917 71 | CUX1_HUMAN.H10MO.C,23707,0.000445933980171,1.87457e-05 72 | BARH1_HUMAN.H10MO.D,27616,0.000429994206146,1.67787e-05 73 | NKX21_HUMAN.H10MO.D,11795,0.0002809035072,-2.74181e-06 74 | HXD4_HUMAN.H10MO.D,19695,0.000339629104047,3.02792e-05 75 | PRRX2_HUMAN.H10MO.C,22715,0.000519275503761,3.16501e-05 76 | PO4F3_HUMAN.H10MO.D,62833,0.00151288356603,0.000109166 77 | HXB6_HUMAN.H10MO.D,29487,0.000832808540549,4.76241e-05 78 | NDF2_HUMAN.H10MO.D,50640,0.00104181998634,1.2666e-05 79 | HXD3_HUMAN.H10MO.D,21962,0.000494465614635,6.16908e-06 80 | SRF_HUMAN.H10MO.A,26754,0.000756134868421,1.52588e-05 81 | ITF2_HUMAN.H10MO.B,21844,0.000561734238764,-2.17557e-05 82 | MYCN_HUMAN.H10MO.B,60757,0.0013033193602,-3.23951e-05 83 | TFE2_HUMAN.H10MO.C,13336,0.00019686843739,2.32458e-06 84 | DLX5_HUMAN.H10MO.D,32288,0.000679971468722,7.37607e-05 85 | PURA_HUMAN.H10MO.D,65532,0.000599823915264,0.000255764 86 | ETV7_HUMAN.H10MO.D,19415,0.000500456168187,-8.22544e-06 87 | NFIA_HUMAN.H10MO.C,28822,0.0007384419407,-1.33216e-05 88 | LHX8_HUMAN.H10MO.D,16909,0.000410909539521,2.19941e-05 89 | COE1_HUMAN.H10MO.A,24472,0.00053140666596,-1.78516e-05 90 | KLF8_HUMAN.H10MO.C,56201,0.0012873076169,6.07073e-05 91 | PEBB_HUMAN.H10MO.C,27349,0.000561094705066,2.71201e-05 92 | STA5A_HUMAN.H10MO.B,8157,0.000207138426611,4.94719e-06 93 | HXD10_HUMAN.H10MO.D,35354,0.00106231253182,0.000125796 94 | ZSC16_HUMAN.H10MO.D,16477,0.000281959083328,5.03659e-06 95 | TBX21_HUMAN.H10MO.D,22388,0.000633537417959,1.15335e-05 96 | GATA5_HUMAN.H10MO.D,13802,0.000234538157248,1.77026e-05 97 | HES1_HUMAN.H10MO.D,57506,0.00188384361966,1.63019e-05 98 | TFCP2_HUMAN.H10MO.D,27935,0.000696205113982,6.97374e-06 99 | TGIF1_HUMAN.H10MO.S,10134,0.000328778824195,-7.80821e-06 100 | BACH1_HUMAN.H10MO.A,17312,0.00046975775913,-1.85966e-05 101 | GSX1_HUMAN.H10MO.D,28750,0.000990368726097,5.32568e-05 102 | TGIF1_HUMAN.H10MO.D,20491,0.000407371920619,9.26852e-06 103 | SOX10_HUMAN.H10MO.D,21909,0.000581212837054,4.04119e-05 104 | RARB_HUMAN.H10MO.D,28100,0.000809726182748,-1.29342e-05 105 | FOXC2_HUMAN.H10MO.D,15490,0.000455181346749,2.17259e-05 106 | NFAC4_HUMAN.H10MO.C,33270,0.00104851310443,-1.93417e-05 107 | TBX5_HUMAN.H10MO.D,15532,0.00056574148778,-5.30481e-06 108 | UNC4_HUMAN.H10MO.D,37148,0.000648857674631,5.83827e-05 109 | MAFG_HUMAN.H10MO.S,9437,0.000311514187869,-1.68681e-05 110 | DRGX_HUMAN.H10MO.D,28773,0.00061168707095,4.58658e-05 111 | BHE41_HUMAN.H10MO.D,37003,0.000991465236793,1.82688e-05 112 | GLIS3_HUMAN.H10MO.D,60568,0.00162449548971,0.000332028 113 | GBX2_HUMAN.H10MO.D,32546,0.000683746693805,5.67436e-05 114 | ELF3_HUMAN.H10MO.D,29286,0.000637125046509,6.04093e-05 115 | TF2LX_HUMAN.H10MO.D,11622,0.00026463725603,2.86102e-06 116 | TEAD3_HUMAN.H10MO.D,16410,0.000371473037128,2.22325e-05 117 | FOXD1_HUMAN.H10MO.D,18675,0.000274404538454,2.30372e-05 118 | CREB1_HUMAN.H10MO.A,31324,0.000721365461661,1.09673e-05 119 | EVX2_HUMAN.H10MO.A,37092,0.00117603622608,6.28233e-05 120 | BSH_HUMAN.H10MO.D,32997,0.000671395702645,3.96073e-05 121 | ZIC3_HUMAN.H10MO.C,67696,0.00197228356169,0.000144809 122 | ISL2_HUMAN.H10MO.D,13895,0.000321537396629,-1.13547e-05 123 | ZIC2_HUMAN.H10MO.C,65926,0.00161737634543,0.000283301 124 | EGR2_HUMAN.H10MO.C,89567,0.0023310975126,0.000422031 125 | FOXQ1_HUMAN.H10MO.C,48263,0.0012584286591,0.000162005 126 | ERR3_HUMAN.H10MO.B,9063,0.000266878343689,-1.38879e-05 127 | GMEB2_HUMAN.H10MO.D,22152,0.00067568705688,-1.11461e-05 128 | JDP2_HUMAN.H10MO.D,26574,0.000630871332178,5.00679e-06 129 | SMAD3_HUMAN.H10MO.C,18717,0.000583993390596,2.97129e-05 130 | ETV5_HUMAN.H10MO.D,17860,0.000397298199501,-2.90573e-05 131 | NKX28_HUMAN.H10MO.C,9413,0.000292462317154,-7.89762e-06 132 | OLIG3_HUMAN.H10MO.D,28981,0.000497477512009,-7.21216e-06 133 | LMX1A_HUMAN.H10MO.D,64077,0.00140434573242,0.00018096 134 | HNF1A_HUMAN.H10MO.A,24786,0.000670352586666,8.76188e-06 135 | NANOG_HUMAN.H10MO.S,25348,0.000469014277546,5.74589e-05 136 | CREM_HUMAN.H10MO.C,31012,0.000633088922631,3.47495e-05 137 | MAFG_HUMAN.H10MO.C,8426,0.000299753971375,-1.69873e-05 138 | PRDM1_HUMAN.H10MO.C,49740,0.000991428707341,1.75536e-05 139 | NANOG_HUMAN.H10MO.A,22875,0.000689007223397,2.59578e-05 140 | ARI5B_HUMAN.H10MO.C,23780,0.000658070220812,5.43892e-05 141 | MEIS2_HUMAN.H10MO.B,14803,0.000467860568091,-1.0699e-05 142 | ZN384_HUMAN.H10MO.C,42887,0.00102384258239,8.32081e-05 143 | PITX2_HUMAN.H10MO.D,25652,0.000621369404821,3.44217e-05 144 | ATF1_HUMAN.H10MO.B,22882,0.000697411603255,2.49743e-05 145 | STAT2_HUMAN.H10MO.B,67818,0.0014599244938,0.000143826 146 | ELK3_HUMAN.H10MO.D,32967,0.000869860504517,-1.41263e-05 147 | NOTO_HUMAN.H10MO.D,20521,0.000421697275036,1.36495e-05 148 | SP4_HUMAN.H10MO.D,77751,0.000854057260907,0.00035724 149 | ZBT18_HUMAN.H10MO.D,16103,0.000304885704497,-5.75185e-06 150 | ELF5_HUMAN.H10MO.D,20484,0.000682584046202,1.04308e-06 151 | NKX61_HUMAN.H10MO.D,46318,0.0012171333449,0.00011602 152 | PKNX1_HUMAN.H10MO.D,17537,0.000671438011613,8.58307e-06 153 | MSX1_HUMAN.H10MO.D,30229,0.000508617781697,0.000104249 154 | DLX4_HUMAN.H10MO.D,39554,0.00100193243474,8.06153e-05 155 | REST_HUMAN.H10MO.A,30055,0.000726930167427,1.39475e-05 156 | RARA_HUMAN.H10MO.S,6166,0.000240294593652,0.0 157 | FOSL1_HUMAN.H10MO.A,10270,0.000245673393923,-4.32134e-06 158 | FUBP1_HUMAN.H10MO.D,53782,0.00147099549808,0.000385493 159 | REL_HUMAN.H10MO.C,29682,0.000504168648434,-4.69983e-05 160 | PO4F1_HUMAN.H10MO.D,58379,0.00123728674475,0.000118762 161 | PHX2A_HUMAN.H10MO.D,28246,0.000442515066028,4.93526e-05 162 | PO2F1_HUMAN.H10MO.B,26132,0.000636397609888,-3.8445e-06 163 | HXB8_HUMAN.H10MO.C,37096,0.00070270679819,6.53565e-05 164 | RFX2_HUMAN.H10MO.C,44360,0.000911929140434,7.75754e-05 165 | VAX1_HUMAN.H10MO.D,29794,0.00104296896976,4.73261e-05 166 | EOMES_HUMAN.H10MO.D,21751,0.000550947386524,3.43323e-05 167 | FOS_HUMAN.H10MO.A,11480,0.000240824775998,-2.26498e-05 168 | ISL1_HUMAN.H10MO.D,20027,0.00064037655958,2.68221e-06 169 | LEF1_HUMAN.H10MO.C,11622,0.00035453193884,1.01328e-05 170 | SRBP2_HUMAN.H10MO.B,80627,0.00177266072539,0.000361234 171 | ATF6A_HUMAN.H10MO.B,83757,0.00178074304965,0.000148058 172 | TFDP1_HUMAN.H10MO.S,52459,0.00136757993754,-1.37091e-05 173 | ZN423_HUMAN.H10MO.D,37167,0.000933377358032,1.24872e-05 174 | NR4A1_HUMAN.H10MO.C,17190,0.000446706681132,7.39098e-06 175 | NR2C2_HUMAN.H10MO.A,44015,0.00111109767447,7.29859e-05 176 | PPARG_HUMAN.H10MO.A,60182,0.00163382756388,0.000107884 177 | TFDP1_HUMAN.H10MO.D,83947,0.00161032021833,0.000401765 178 | RFX4_HUMAN.H10MO.D,20241,0.000389572860089,2.20537e-05 179 | CEBPG_HUMAN.H10MO.C,32488,0.000453876295271,-1.28448e-05 180 | NFAC1_HUMAN.H10MO.S,18602,0.000458166715482,-3.76701e-05 181 | PPARG_HUMAN.H10MO.S,18175,0.000633442949263,8.13603e-06 182 | FOXI1_HUMAN.H10MO.B,23741,0.000521076197532,7.92742e-06 183 | TBX1_HUMAN.H10MO.D,48409,0.00069040054183,0.000299931 184 | SOX7_HUMAN.H10MO.D,22185,0.000517161061956,2.94745e-05 185 | GATA1_HUMAN.H10MO.S,13754,0.000301381203865,5.0962e-06 186 | MYOD1_HUMAN.H10MO.C,19364,0.000452304295573,-2.06232e-05 187 | WT1_HUMAN.H10MO.D,96852,0.00238263622849,0.000580907 188 | BHE22_HUMAN.H10MO.D,41646,0.000521392690362,5.1707e-05 189 | GATA1_HUMAN.H10MO.A,20124,0.000461033155188,4.11272e-06 190 | ERG_HUMAN.H10MO.B,33606,0.000773325110069,-6.25849e-05 191 | E2F7_HUMAN.H10MO.D,47647,0.00114401560075,3.06964e-06 192 | HAND1_HUMAN.H10MO.D,11538,0.000291380980746,-1.03712e-05 193 | NKX31_HUMAN.H10MO.C,49373,0.00109436314744,0.000159979 194 | HXD8_HUMAN.H10MO.D,21662,0.000609633799423,-4.23193e-06 195 | ELK1_HUMAN.H10MO.A,23895,0.000699316126332,-2.82526e-05 196 | FOSB_HUMAN.H10MO.C,11277,0.000251016951345,9.53674e-07 197 | TBX2_HUMAN.H10MO.D,14059,0.000319850066434,4.05312e-06 198 | ESR1_HUMAN.H10MO.A,24848,0.000462662850246,-8.9407e-06 199 | RUNX3_HUMAN.H10MO.C,26104,0.000720703269161,2.69413e-05 200 | NKX23_HUMAN.H10MO.D,16022,0.000481273899244,1.81198e-05 201 | NKX62_HUMAN.H10MO.D,26371,0.000720985736421,2.32458e-06 202 | ESR1_HUMAN.H10MO.S,25358,0.000407059477424,8.61287e-06 203 | MAZ_HUMAN.H10MO.A,95202,0.0010708936664,0.000456601 204 | HXA7_HUMAN.H10MO.D,10894,0.000306014454986,1.12951e-05 205 | CRX_HUMAN.H10MO.C,14586,0.000367268940188,3.96371e-06 206 | SHOX_HUMAN.H10MO.D,39451,0.000973940269169,0.000107467 207 | NOBOX_HUMAN.H10MO.C,17412,0.000327906837334,3.8147e-06 208 | MZF1_HUMAN.H10MO.D,70496,0.00170346165019,0.000241846 209 | STAT1_HUMAN.H10MO.A,20143,0.000410082475367,-1.39475e-05 210 | CTCFL_HUMAN.H10MO.A,49709,0.00112202918427,8.59797e-05 211 | CEBPD_HUMAN.H10MO.B,33712,0.00071420708562,3.8147e-06 212 | TF7L2_HUMAN.H10MO.A,22013,0.000599081490396,2.0057e-05 213 | ID4_HUMAN.H10MO.D,23578,0.000460713848749,-1.88947e-05 214 | PAX3_HUMAN.H10MO.D,27749,0.000461632620381,1.66595e-05 215 | STAT1_HUMAN.H10MO.S,14373,0.000283060780713,-3.23355e-05 216 | DLX1_HUMAN.H10MO.D,32138,0.000844692624037,5.9098e-05 217 | PAX7_HUMAN.H10MO.D,29777,0.000554559926375,2.48253e-05 218 | NKX22_HUMAN.H10MO.D,11139,0.000253873131597,1.35601e-05 219 | RXRG_HUMAN.H10MO.B,21779,0.00056651859085,7.21216e-06 220 | TBX20_HUMAN.H10MO.D,62592,0.00177775995249,0.000222892 221 | MUSC_HUMAN.H10MO.D,33449,0.000693140652862,5.96046e-08 222 | ZSCA4_HUMAN.H10MO.D,32412,0.000865068824812,3.32892e-05 223 | VSX2_HUMAN.H10MO.D,10973,0.000193902058403,1.69873e-06 224 | NFAT5_HUMAN.H10MO.D,19065,0.000184434931617,-2.75373e-05 225 | BCL6_HUMAN.H10MO.C,6873,0.000147481661658,1.055e-05 226 | ZIC1_HUMAN.H10MO.B,66122,0.00183856096779,0.000271112 227 | MEF2D_HUMAN.H10MO.C,18040,0.000364257789985,3.68655e-05 228 | SP3_HUMAN.H10MO.B,65556,0.000730947454289,0.000293702 229 | NR1H4_HUMAN.H10MO.C,28509,0.000441370596816,4.08888e-05 230 | E2F4_HUMAN.H10MO.A,65253,0.00193894552839,-1.96397e-05 231 | E2F5_HUMAN.H10MO.B,53043,0.00101431941867,1.94609e-05 232 | HLF_HUMAN.H10MO.C,31143,0.000587768523321,-1.92225e-05 233 | TBX4_HUMAN.H10MO.D,19632,0.000173841745924,1.17421e-05 234 | HXD13_HUMAN.H10MO.D,15740,0.000439343511172,2.70903e-05 235 | HME1_HUMAN.H10MO.D,31909,0.000538337266498,7.4774e-05 236 | TYY1_HUMAN.H10MO.A,33005,0.000824037987306,-1.68085e-05 237 | FOXF1_HUMAN.H10MO.D,63738,0.00176068467383,0.000278741 238 | EMX1_HUMAN.H10MO.D,25201,0.000513881403242,2.71797e-05 239 | ZN232_HUMAN.H10MO.D,20029,0.000610929459235,1.46925e-05 240 | BC11A_HUMAN.H10MO.C,69459,0.00214309149697,0.000114799 241 | SOX13_HUMAN.H10MO.D,41830,0.00110389401744,0.000108629 242 | NKX32_HUMAN.H10MO.C,23041,0.000578535854937,-6.67572e-06 243 | CREB3_HUMAN.H10MO.D,60842,0.00106091439447,6.85155e-05 244 | PO2F3_HUMAN.H10MO.D,21028,0.000501395917092,-7.56979e-06 245 | HMX2_HUMAN.H10MO.D,21444,0.000331340163463,4.59552e-05 246 | SNAI1_HUMAN.H10MO.C,24141,0.000634692781867,-3.8892e-05 247 | HXC6_HUMAN.H10MO.D,50236,0.00137655684244,0.000296384 248 | P63_HUMAN.H10MO.S,16594,0.000210171800076,-7.62939e-06 249 | VAX2_HUMAN.H10MO.D,18118,0.000635892212134,1.76728e-05 250 | TBX19_HUMAN.H10MO.D,20338,0.000432612763288,9.59635e-06 251 | TLX1_HUMAN.H10MO.S,27704,0.000644610741258,5.95748e-05 252 | BATF3_HUMAN.H10MO.D,32454,0.000786741468489,-4.88758e-06 253 | HXB1_HUMAN.H10MO.D,26808,0.000605627561077,2.6077e-05 254 | PDX1_HUMAN.H10MO.C,10656,0.000259352410137,6.34789e-06 255 | TF65_HUMAN.H10MO.A,26720,0.000539597657903,-5.26309e-05 256 | LHX6_HUMAN.H10MO.D,22755,0.000477814068216,4.15742e-05 257 | E4F1_HUMAN.H10MO.D,47755,0.00116145289914,-1.63913e-06 258 | BHE23_HUMAN.H10MO.D,47845,0.00119896537675,6.59823e-05 259 | SOX3_HUMAN.H10MO.D,23600,0.000666777952982,2.46763e-05 260 | NGN2_HUMAN.H10MO.D,41150,0.00103006073819,-2.94149e-05 261 | NFYB_HUMAN.H10MO.A,38542,0.000562899741051,7.91252e-05 262 | ALX4_HUMAN.H10MO.D,31040,0.000759474016376,6.02305e-05 263 | MAFK_HUMAN.H10MO.A,8201,0.000250880507967,-1.04904e-05 264 | HME2_HUMAN.H10MO.D,18961,0.000348353955824,8.91089e-06 265 | BARX1_HUMAN.H10MO.D,33237,0.000874454289142,9.92119e-05 266 | HXC11_HUMAN.H10MO.D,24271,0.000530884581195,-1.48416e-05 267 | HXB3_HUMAN.H10MO.D,28830,0.000754238099034,4.97699e-05 268 | ATF7_HUMAN.H10MO.D,25681,0.000563091099574,-1.0699e-05 269 | OLIG1_HUMAN.H10MO.D,31466,0.000716692652758,1.93119e-05 270 | CR3L2_HUMAN.H10MO.D,80573,0.00188824197207,9.53972e-05 271 | JUND_HUMAN.H10MO.A,11063,0.00030153175077,-8.10623e-06 272 | CDC5L_HUMAN.H10MO.D,33905,0.000830519951558,1.21593e-05 273 | MEOX1_HUMAN.H10MO.D,24611,0.000677450910141,2.27988e-05 274 | EPAS1_HUMAN.H10MO.D,81856,0.00108228471183,0.000259936 275 | CEBPB_HUMAN.H10MO.A,26793,0.000625650337668,-1.2815e-05 276 | IRF3_HUMAN.H10MO.C,54457,0.00125055789419,9.59337e-05 277 | KAISO_HUMAN.H10MO.A,71373,0.00166276698883,0.00024426 278 | RUNX1_HUMAN.H10MO.A,23805,0.000607610265685,2.22325e-05 279 | PO3F3_HUMAN.H10MO.D,69345,0.00150574666968,0.000268608 280 | IRX2_HUMAN.H10MO.D,73509,0.00189073991764,0.000287592 281 | KAISO_HUMAN.H10MO.S,16350,0.000393793798469,2.08616e-06 282 | MYB_HUMAN.H10MO.C,26287,0.000508499310013,2.44081e-05 283 | GFI1_HUMAN.H10MO.C,13365,0.000256488253283,1.15931e-05 284 | ETV4_HUMAN.H10MO.B,14059,0.000414193730425,1.2219e-06 285 | NR5A2_HUMAN.H10MO.C,10323,0.000110015454841,1.46925e-05 286 | HBP1_HUMAN.H10MO.D,16637,0.000484082027444,1.23382e-05 287 | ARX_HUMAN.H10MO.D,26364,0.000563092661959,5.40614e-05 288 | EMX2_HUMAN.H10MO.D,27173,0.000682534083027,3.29316e-05 289 | ARNT2_HUMAN.H10MO.D,64429,0.00144476666104,0.00022167 290 | ZN333_HUMAN.H10MO.D,45117,0.00107609624648,0.000107467 291 | NR2E1_HUMAN.H10MO.D,21161,0.000551712281912,2.56598e-05 292 | FOXA3_HUMAN.H10MO.C,49131,0.00114485235505,0.000224441 293 | BARX2_HUMAN.H10MO.D,29525,0.000879059674604,4.50611e-05 294 | TEF_HUMAN.H10MO.D,28131,0.000977833337508,5.08726e-05 295 | ONEC3_HUMAN.H10MO.D,24821,0.000710720557013,8.75294e-05 296 | ONEC2_HUMAN.H10MO.D,40504,0.00108526295414,0.000177771 297 | ZN282_HUMAN.H10MO.D,54532,0.00129335115004,1.8239e-05 298 | HXB13_HUMAN.H10MO.D,22776,0.000515269544789,7.53999e-06 299 | MIXL1_HUMAN.H10MO.D,19262,0.0003444079817,1.43051e-05 300 | FOXP2_HUMAN.H10MO.A,24108,0.000561999253124,3.03686e-05 301 | FOXJ3_HUMAN.H10MO.S,50190,0.000968902089429,0.000362754 302 | TWST1_HUMAN.H10MO.D,29685,0.000728207754218,-1.54078e-05 303 | OTX1_HUMAN.H10MO.D,7698,0.000209527783751,-6.10948e-06 304 | NR6A1_HUMAN.H10MO.B,16530,0.000480665379259,2.41399e-06 305 | HXD9_HUMAN.H10MO.D,25673,0.000669488003518,3.84152e-05 306 | BRAC_HUMAN.H10MO.D,13603,0.000236259182859,-2.11e-05 307 | ERF_HUMAN.H10MO.D,24950,0.000629771985763,3.66569e-06 308 | FOXA1_HUMAN.H10MO.A,24019,0.000588365214709,3.56138e-05 309 | FOXJ3_HUMAN.H10MO.A,57903,0.00145756868085,0.000546008 310 | MYOG_HUMAN.H10MO.D,12317,0.000248995558383,-1.31726e-05 311 | BARH2_HUMAN.H10MO.D,27248,0.00052423434581,1.4931e-05 312 | MAFB_HUMAN.H10MO.D,12198,0.000169444470873,-1.96397e-05 313 | GLI1_HUMAN.H10MO.C,49361,0.0011813017891,0.000122756 314 | TLX1_HUMAN.H10MO.D,19559,0.000359054009025,-5.96046e-07 315 | PLAG1_HUMAN.H10MO.D,90223,0.000878814232935,0.000395507 316 | E2F1_HUMAN.H10MO.A,50424,0.00134965441192,5.01871e-05 317 | YBOX1_HUMAN.H10MO.D,34672,0.00102439624263,-1.5527e-05 318 | IRF1_HUMAN.H10MO.A,98628,0.00205738866993,0.000185817 319 | HMX1_HUMAN.H10MO.D,29693,0.000785974875919,7.59065e-05 320 | SOX1_HUMAN.H10MO.D,17852,0.000529993742673,2.19047e-05 321 | TBX15_HUMAN.H10MO.D,78471,0.000679335499794,0.000314564 322 | PPARD_HUMAN.H10MO.D,27895,0.000434800351696,2.0057e-05 323 | RFX3_HUMAN.H10MO.B,15120,0.000297186216546,-3.42727e-06 324 | USF2_HUMAN.H10MO.A,62402,0.00123438216719,5.42402e-06 325 | RREB1_HUMAN.H10MO.D,71139,0.000772402000546,0.000281453 326 | MAFF_HUMAN.H10MO.A,8224,0.000140410309379,-1.49012e-06 327 | FOXD2_HUMAN.H10MO.D,41701,0.000877725033881,0.000101805 328 | E2F3_HUMAN.H10MO.B,72092,0.00189645204295,4.01735e-05 329 | PO4F2_HUMAN.H10MO.D,22938,0.000581281097684,1.72257e-05 330 | THA_HUMAN.H10MO.S,15455,0.000271901250098,-3.27826e-05 331 | EGR4_HUMAN.H10MO.D,78038,0.001772051089,0.000453204 332 | HIC2_HUMAN.H10MO.D,33132,0.000869129573353,-1.64807e-05 333 | NFIA_HUMAN.H10MO.S,4833,5.74798065593e-05,-3.30806e-06 334 | STAT4_HUMAN.H10MO.D,11476,0.000139310430354,3.66569e-06 335 | TGIF2_HUMAN.H10MO.D,12561,0.00033479839818,-1.57654e-05 336 | MCR_HUMAN.H10MO.D,18773,0.000265587825407,-5.51343e-06 337 | PITX3_HUMAN.H10MO.D,23152,0.000695433795159,1.2666e-05 338 | SHOX2_HUMAN.H10MO.D,38283,0.000975450924686,9.26852e-05 339 | NR1D1_HUMAN.H10MO.C,16471,0.000595281356169,1.35303e-05 340 | MEIS1_HUMAN.H10MO.C,12963,0.000305993122176,0.0 341 | ZN740_HUMAN.H10MO.D,42542,0.00125348171002,0.000428796 342 | PRDM4_HUMAN.H10MO.D,23935,0.000525878626912,2.27392e-05 343 | GSC_HUMAN.H10MO.D,24811,0.00057985539311,-1.07884e-05 344 | DMBX1_HUMAN.H10MO.D,36926,0.000790162105861,6.93798e-05 345 | PO6F2_HUMAN.H10MO.D,11231,0.000352617764934,-3.51667e-06 346 | SOX5_HUMAN.H10MO.C,51610,0.00129175547071,0.000162125 347 | NRL_HUMAN.H10MO.D,14188,0.000452761849107,-1.42753e-05 348 | ATF3_HUMAN.H10MO.A,21830,0.000660341705897,2.10404e-05 349 | SP1_HUMAN.H10MO.C,55649,0.000808965948931,0.000269681 350 | ETV6_HUMAN.H10MO.D,35727,0.00078009288551,0.000114232 351 | IRF8_HUMAN.H10MO.D,60331,0.00140706683932,3.5733e-05 352 | SP1_HUMAN.H10MO.S,56311,0.00167043934917,0.000250489 353 | SPZ1_HUMAN.H10MO.D,86905,0.00199537290615,0.000336379 354 | BPTF_HUMAN.H10MO.D,57698,0.00154082165009,0.000405401 355 | PHX2B_HUMAN.H10MO.D,28561,0.00068331666952,4.36604e-05 356 | HSF1_HUMAN.H10MO.A,13439,0.000358624342408,-2.40505e-05 357 | CXXC1_HUMAN.H10MO.D,31035,0.000776716441711,-6.4373e-06 358 | FOXO3_HUMAN.H10MO.B,45746,0.00113323713607,0.000190943 359 | MYBB_HUMAN.H10MO.D,21352,0.000448663343319,4.3571e-05 360 | GATA6_HUMAN.H10MO.B,17164,0.000371962220922,1.63913e-06 361 | FOXK1_HUMAN.H10MO.D,87649,0.00186355169452,0.000276983 362 | BMAL1_HUMAN.H10MO.C,50350,0.00114986386772,-2.92361e-05 363 | GLI2_HUMAN.H10MO.B,33768,0.000898646369803,5.67436e-05 364 | TEAD1_HUMAN.H10MO.D,22895,0.00073151610688,-8.13603e-06 365 | MAFK_HUMAN.H10MO.S,7968,0.000355967118905,-4.14252e-06 366 | HXD11_HUMAN.H10MO.D,35943,0.000948509163328,6.78599e-05 367 | TFE3_HUMAN.H10MO.C,37881,0.000926510888287,-1.45435e-05 368 | TYY2_HUMAN.H10MO.D,56667,0.00175356118561,-8.9407e-08 369 | HIF1A_HUMAN.H10MO.A,78191,0.00159779266667,1.07884e-05 370 | NKX25_HUMAN.H10MO.C,6526,0.000263598036618,1.49012e-07 371 | KLF14_HUMAN.H10MO.D,71824,0.00199834055968,0.000314176 372 | NDF1_HUMAN.H10MO.C,31628,0.000643862479734,7.80821e-06 373 | NR0B1_HUMAN.H10MO.D,44620,0.00133188030263,6.98864e-05 374 | LHX9_HUMAN.H10MO.D,36773,0.00100097274277,9.97782e-05 375 | HXA10_HUMAN.H10MO.C,31033,0.000877345469338,6.25849e-05 376 | PTF1A_HUMAN.H10MO.C,25112,0.000749342521739,-7.53999e-06 377 | NF2L1_HUMAN.H10MO.C,9437,0.000200509044346,-2.68817e-05 378 | ATOH1_HUMAN.H10MO.D,28484,0.000760154091564,-3.01003e-06 379 | ZBED1_HUMAN.H10MO.D,19742,0.000298528238333,-3.67761e-05 380 | GATA4_HUMAN.H10MO.B,13065,0.000320074860476,7.86781e-06 381 | SOX9_HUMAN.H10MO.B,42283,0.00141571627303,6.14226e-05 382 | LHX4_HUMAN.H10MO.D,28821,0.000517081980996,2.8342e-05 383 | NR4A2_HUMAN.H10MO.C,15541,0.000336880459302,-3.30806e-06 384 | MNX1_HUMAN.H10MO.D,41411,0.000725101445934,6.45518e-05 385 | MGAP_HUMAN.H10MO.D,19218,0.000219588905946,-1.05798e-05 386 | AP2D_HUMAN.H10MO.D,73517,0.00146686569858,9.05097e-05 387 | BCL6B_HUMAN.H10MO.D,8970,0.000216162332805,-3.75509e-06 388 | HXA11_HUMAN.H10MO.D,33181,0.00114146193124,5.08726e-05 389 | SMRC1_HUMAN.H10MO.D,12187,0.000220781359442,-1.20401e-05 390 | HMGA1_HUMAN.H10MO.D,25968,0.000855641998477,2.7597e-05 391 | JUNB_HUMAN.H10MO.C,12620,0.000214294961399,-5.84126e-06 392 | P53_HUMAN.H10MO.B,20271,0.000334577797805,1.09375e-05 393 | FOXA2_HUMAN.H10MO.A,24834,0.000641148923583,7.04825e-05 394 | IRF5_HUMAN.H10MO.D,93217,0.00221129769085,0.000379264 395 | NFYC_HUMAN.H10MO.B,56222,0.00118765442928,0.000134736 396 | ARNT_HUMAN.H10MO.B,61873,0.00132603366299,2.05338e-05 397 | RORG_HUMAN.H10MO.C,19344,0.00046410472988,-3.18885e-06 398 | RAX2_HUMAN.H10MO.D,31453,0.00067813132886,6.82473e-05 399 | COT1_HUMAN.H10MO.B,13788,0.000276013723824,1.69873e-06 400 | THB_HUMAN.H10MO.S,16626,0.000365922037177,-2.68817e-05 401 | ZEB1_HUMAN.H10MO.B,21566,0.000638273162938,-2.88785e-05 402 | SMAD4_HUMAN.H10MO.C,20215,0.000636061325277,4.76837e-06 403 | THB_HUMAN.H10MO.C,27331,0.000764047586803,2.29776e-05 404 | KLF12_HUMAN.H10MO.D,80652,0.00176904382623,0.000173151 405 | SOX21_HUMAN.H10MO.D,21016,0.000609174848528,2.25604e-05 406 | E2F6_HUMAN.H10MO.C,90064,0.00195836703983,0.000398487 407 | MBD2_HUMAN.H10MO.B,61855,0.00117475293184,8.55625e-05 408 | SOX2_HUMAN.H10MO.B,26763,0.000597705331142,1.93119e-05 409 | P73_HUMAN.H10MO.S,25484,0.000504522515083,2.5928e-06 410 | SMAD2_HUMAN.H10MO.C,18534,0.000738122088515,-1.13249e-06 411 | EGR1_HUMAN.H10MO.A,71249,0.00172174932071,0.000418872 412 | JUN_HUMAN.H10MO.A,10841,0.000168544628576,-6.25849e-06 413 | CEBPE_HUMAN.H10MO.A,29659,0.000584062139579,4.44949e-05 414 | RARG_HUMAN.H10MO.C,39630,0.00119917730084,6.02007e-06 415 | IRX3_HUMAN.H10MO.D,40777,0.00106317512479,9.19104e-05 416 | EGR3_HUMAN.H10MO.D,37479,0.000773447911118,8.90493e-05 417 | P73_HUMAN.H10MO.A,22775,0.000533776399431,-1.2517e-05 418 | HNF6_HUMAN.H10MO.C,22523,0.000599809083774,4.13358e-05 419 | DUXA_HUMAN.H10MO.D,14587,0.000420702903796,7.24196e-06 420 | FOXL1_HUMAN.H10MO.D,90843,0.00150407848502,0.00056538 421 | EGR1_HUMAN.H10MO.S,57011,0.00124303796997,0.000231594 422 | HLTF_HUMAN.H10MO.D,21135,0.000620789167673,-2.72989e-05 423 | RARG_HUMAN.H10MO.S,6166,0.000285303207452,1.75834e-06 424 | COT2_HUMAN.H10MO.A,27818,0.000485839290157,3.66271e-05 425 | UBIP1_HUMAN.H10MO.D,7799,0.000360155730427,5.126e-06 426 | MAF_HUMAN.H10MO.B,9087,0.000147929943677,-1.37389e-05 427 | HXC13_HUMAN.H10MO.D,23213,0.000845554945603,-8.9407e-06 428 | DDIT3_HUMAN.H10MO.C,28635,0.000484438604624,3.96073e-05 429 | CPEB1_HUMAN.H10MO.D,40346,0.000829186835518,0.000330776 430 | COT2_HUMAN.H10MO.S,35081,0.00104041748867,1.85966e-05 431 | PITX1_HUMAN.H10MO.D,27383,0.000478664671798,2.82526e-05 432 | ENOA_HUMAN.H10MO.A,59669,0.0012195532039,-3.28124e-05 433 | ZKSC3_HUMAN.H10MO.D,5744,0.000112866653406,-8.37445e-06 434 | RXRB_HUMAN.H10MO.C,18200,0.000470520797559,-3.62098e-05 435 | NFE2_HUMAN.H10MO.B,11790,0.000416426506596,4.11272e-06 436 | TAL1_HUMAN.H10MO.A,17222,0.000113767149408,7.80821e-06 437 | SOX11_HUMAN.H10MO.D,20802,0.000595102689562,3.21567e-05 438 | GSX2_HUMAN.H10MO.D,35275,0.000763220774313,7.81715e-05 439 | TAL1_HUMAN.H10MO.S,26828,0.000369962314853,-6.34789e-06 440 | BRCA1_HUMAN.H10MO.D,39296,0.000650325987358,0.000106663 441 | FEV_HUMAN.H10MO.C,20915,0.000616431408105,-3.62098e-05 442 | VSX1_HUMAN.H10MO.D,32349,0.000717845228478,5.80251e-05 443 | PBX3_HUMAN.H10MO.B,19343,0.000432314473507,3.6031e-05 444 | MLX_HUMAN.H10MO.D,45200,0.000768540025469,-4.3124e-05 445 | IRF7_HUMAN.H10MO.C,46770,0.000931519279918,4.98295e-05 446 | PO3F1_HUMAN.H10MO.C,24804,0.000347330692051,-3.39746e-06 447 | SRY_HUMAN.H10MO.B,77970,0.0015903373716,0.000345588 448 | HMX3_HUMAN.H10MO.D,20735,0.000188800325661,2.41101e-05 449 | SOX4_HUMAN.H10MO.C,39426,0.00113510873588,4.25279e-05 450 | CDX1_HUMAN.H10MO.C,27688,0.000846373662132,3.79682e-05 451 | RARA_HUMAN.H10MO.C,42044,0.00120831525028,5.83827e-05 452 | STF1_HUMAN.H10MO.B,13698,0.000208759918974,1.67191e-05 453 | HMBX1_HUMAN.H10MO.D,6511,0.000210444288792,-5.27501e-06 454 | ZN652_HUMAN.H10MO.D,37486,0.000796259227711,5.48661e-05 455 | MECP2_HUMAN.H10MO.C,33192,0.000748659139074,-2.55108e-05 456 | VDR_HUMAN.H10MO.B,65400,0.00152475223626,0.00055021 457 | ZBT49_HUMAN.H10MO.D,19103,0.00033605009387,-5.87106e-06 458 | AP2C_HUMAN.H10MO.A,43260,0.000743086105285,3.74317e-05 459 | RFX1_HUMAN.H10MO.C,20585,0.000771385169594,-1.81198e-05 460 | NFAC3_HUMAN.H10MO.B,26405,0.000426756911874,-3.15905e-06 461 | NFAC1_HUMAN.H10MO.A,29601,0.000844603842003,-3.69549e-05 462 | VDR_HUMAN.H10MO.S,21764,0.00076244011657,5.60284e-06 463 | NR2E3_HUMAN.H10MO.C,40946,0.000724715891348,0.000238091 464 | VENTX_HUMAN.H10MO.D,24123,0.000480274575924,2.90573e-05 465 | SOX15_HUMAN.H10MO.D,26475,0.00073258216467,2.3216e-05 466 | NR1I3_HUMAN.H10MO.C,17317,0.000528959213312,9.38773e-06 467 | CREB5_HUMAN.H10MO.D,23908,0.000472248451221,8.67248e-06 468 | RHXF1_HUMAN.H10MO.D,27188,0.000783427332287,-1.33216e-05 469 | TCF7_HUMAN.H10MO.C,17781,0.000315249934447,1.95801e-05 470 | HOMEZ_HUMAN.H10MO.D,59704,0.00118569143686,8.11815e-05 471 | SUH_HUMAN.H10MO.C,23688,0.000527262271261,2.11596e-06 472 | NR1I3_HUMAN.H10MO.S,12088,0.000147006911355,-7.15256e-06 473 | GABP1_HUMAN.H10MO.C,47486,0.00117522887683,8.66354e-05 474 | KLF4_HUMAN.H10MO.A,96874,0.00225764963348,0.000344813 475 | TBX3_HUMAN.H10MO.D,16221,0.000230589894521,1.75834e-06 476 | GLIS1_HUMAN.H10MO.D,55576,0.00138712315072,0.000280797 477 | FLI1_HUMAN.H10MO.A,51812,0.00121209908328,0.000144184 478 | DLX3_HUMAN.H10MO.C,16097,0.000178045199746,1.84774e-06 479 | DLX6_HUMAN.H10MO.D,41913,0.00104428854601,9.62019e-05 480 | GATA2_HUMAN.H10MO.A,17026,0.000433697589768,4.76837e-06 481 | CENPB_HUMAN.H10MO.D,59064,0.00129457671985,0.000127614 482 | CLOCK_HUMAN.H10MO.D,65852,0.000728912644645,0.000318229 483 | ALX1_HUMAN.H10MO.B,29968,0.000595919410276,8.70228e-05 484 | GLI3_HUMAN.H10MO.B,57190,0.00142831830131,0.000153482 485 | FOXG1_HUMAN.H10MO.D,87596,0.0014258704474,0.000662506 486 | HEY1_HUMAN.H10MO.D,58559,0.00109916006333,-3.90708e-05 487 | ZN219_HUMAN.H10MO.D,51709,0.0014008532498,0.000443071 488 | BHE40_HUMAN.H10MO.A,45114,0.000846934741071,-3.14116e-05 489 | MEF2B_HUMAN.H10MO.D,13104,0.000440086265664,3.33786e-06 490 | HMGA2_HUMAN.H10MO.D,39976,0.00113903055515,-1.02222e-05 491 | IRF2_HUMAN.H10MO.C,58757,0.00148836810345,4.87864e-05 492 | MEF2A_HUMAN.H10MO.A,16205,0.000246709550587,4.74751e-05 493 | ZBT7B_HUMAN.H10MO.D,39020,0.000731494912285,7.17342e-05 494 | HES5_HUMAN.H10MO.D,60797,0.00140316011702,-4.59552e-05 495 | HXA5_HUMAN.H10MO.D,32402,0.000825424186113,4.12464e-05 496 | ISX_HUMAN.H10MO.D,31466,0.00064677734125,7.98106e-05 497 | PAX5_HUMAN.H10MO.A,66198,0.00175037622847,8.50558e-05 498 | HES7_HUMAN.H10MO.D,50575,0.0011098706625,-5.37634e-05 499 | HNF4A_HUMAN.H10MO.A,19859,0.000248974987395,-2.0802e-05 500 | NR2C1_HUMAN.H10MO.C,16400,0.000346031383255,-1.16229e-06 501 | GBX1_HUMAN.H10MO.D,36446,0.000885702410149,9.01222e-05 502 | FOXC1_HUMAN.H10MO.C,28553,0.000878306526205,6.81281e-05 503 | MITF_HUMAN.H10MO.C,30654,0.000577701487136,-3.79086e-05 504 | PAX5_HUMAN.H10MO.S,6397,2.93677000099e-05,-3.01003e-06 505 | COT1_HUMAN.H10MO.S,19871,0.000510695936615,-1.11163e-05 506 | NFAC2_HUMAN.H10MO.B,33346,0.000803202966247,-3.66271e-05 507 | PKNX2_HUMAN.H10MO.D,56070,0.00137041600303,0.000286222 508 | OTX2_HUMAN.H10MO.C,24104,0.000659225554428,1.71065e-05 509 | MEF2C_HUMAN.H10MO.C,24836,0.000740118163443,0.00011754 510 | NR1I2_HUMAN.H10MO.C,16819,0.000358850235109,2.44379e-06 511 | KLF16_HUMAN.H10MO.D,65795,0.00102869591358,0.000343114 512 | ZN589_HUMAN.H10MO.D,64214,0.00109499185092,0.000259519 513 | HXA1_HUMAN.H10MO.C,28137,0.000968070985683,5.87702e-05 514 | TBR1_HUMAN.H10MO.D,25842,0.000602709918251,2.13683e-05 515 | GABPA_HUMAN.H10MO.A,50526,0.001369142397,0.000158221 516 | ELK4_HUMAN.H10MO.A,37692,0.000694140626083,0.00010246 517 | IRF9_HUMAN.H10MO.C,26849,0.000613881281661,6.82473e-06 518 | FOXO6_HUMAN.H10MO.D,27922,0.000687142450117,3.98755e-05 519 | DPRX_HUMAN.H10MO.D,24316,0.000493036537206,1.2219e-06 520 | EVI1_HUMAN.H10MO.B,23256,0.000736652627005,1.22786e-05 521 | NFKB1_HUMAN.H10MO.B,29536,0.000536422408018,-3.39448e-05 522 | EHF_HUMAN.H10MO.S,29416,0.000768437941639,4.47035e-06 523 | TFAP4_HUMAN.H10MO.C,22446,0.00048986377512,2.04742e-05 524 | ZKSC1_HUMAN.H10MO.C,32384,0.0006692877521,1.33812e-05 525 | CTCF_HUMAN.H10MO.A,53493,0.00094898410624,8.06749e-05 526 | IKZF1_HUMAN.H10MO.C,24254,0.000617447078067,2.87294e-05 527 | HEN1_HUMAN.H10MO.C,30831,0.00073448355548,2.86102e-06 528 | E2F2_HUMAN.H10MO.B,71465,0.00196834374786,-4.14252e-05 529 | EHF_HUMAN.H10MO.C,87085,0.0019120533166,8.02577e-05 530 | PROP1_HUMAN.H10MO.D,24712,0.000365285333754,2.15173e-05 531 | KLF3_HUMAN.H10MO.D,49921,0.000999889420204,0.000163168 532 | SNAI2_HUMAN.H10MO.C,15531,0.000265171956798,-2.56002e-05 533 | TBP_HUMAN.H10MO.C,29589,0.00056915298436,4.8548e-05 534 | PAX8_HUMAN.H10MO.D,23559,0.000586729489686,-4.60446e-05 535 | GCR_HUMAN.H10MO.S,9432,0.000272810715301,1.54972e-06 536 | ZN350_HUMAN.H10MO.C,41925,0.000982704835295,9.10163e-05 537 | HINFP_HUMAN.H10MO.C,36104,0.000862501391008,-7.45058e-07 538 | HXC12_HUMAN.H10MO.D,30138,0.00106069086102,2.77758e-05 539 | ZN143_HUMAN.H10MO.A,70064,0.00184240750799,0.000139445 540 | NR4A3_HUMAN.H10MO.D,13940,0.000445399610364,3.27826e-07 541 | HESX1_HUMAN.H10MO.D,30765,0.000847259816566,-2.80738e-05 542 | PBX1_HUMAN.H10MO.B,23515,0.000652738049892,3.45111e-05 543 | PRGR_HUMAN.H10MO.C,42693,0.00103680390518,0.000217646 544 | GCR_HUMAN.H10MO.A,47881,0.00138961285363,0.000203937 545 | NFIL3_HUMAN.H10MO.C,24458,0.000513989519304,-8.79169e-06 546 | SMAD1_HUMAN.H10MO.D,29337,0.000889961561244,2.61962e-05 547 | HXB7_HUMAN.H10MO.C,37073,0.000875646714105,5.24223e-05 548 | PAX4_HUMAN.H10MO.D,26438,0.000487764401269,5.06341e-05 549 | SPIB_HUMAN.H10MO.B,54294,0.00145405917002,8.01682e-06 550 | PPARA_HUMAN.H10MO.S,7287,0.000294479760295,-4.05312e-06 551 | CEBPA_HUMAN.H10MO.A,20278,0.000564426799049,6.25849e-06 552 | HXA9_HUMAN.H10MO.D,18246,0.00051884915911,3.15607e-05 553 | HIC1_HUMAN.H10MO.C,40296,0.0010815826164,-4.68493e-05 554 | THAP1_HUMAN.H10MO.D,93697,0.00181919028003,0.000301957 555 | PPARA_HUMAN.H10MO.C,34531,0.000726972757984,2.50936e-05 556 | FOXO1_HUMAN.H10MO.C,85177,0.00161447957556,0.00058651 557 | GCM1_HUMAN.H10MO.D,35911,0.000765615910292,-1.67191e-05 558 | CUX2_HUMAN.H10MO.D,21497,0.000246537345192,3.53754e-05 559 | RORA_HUMAN.H10MO.B,15804,0.0004873718346,-9.35793e-06 560 | ZBT7A_HUMAN.H10MO.D,72813,0.00166627219713,0.000122994 561 | NR1I2_HUMAN.H10MO.S,12279,0.000203991010056,-9.26852e-06 562 | USF1_HUMAN.H10MO.A,40417,0.000850378949418,-1.63019e-05 563 | HXC10_HUMAN.H10MO.D,70296,0.00198166310914,0.00033012 564 | SOX18_HUMAN.H10MO.D,31587,0.000867563391947,-2.6226e-06 565 | LMX1B_HUMAN.H10MO.D,63339,0.00104733158891,0.000187188 566 | CEBPZ_HUMAN.H10MO.D,34714,0.000847907613137,9.65893e-05 567 | ZN410_HUMAN.H10MO.D,36684,0.000924578195873,-3.96967e-05 568 | EVX1_HUMAN.H10MO.D,55696,0.000844633431561,0.000195056 569 | P5F1B_HUMAN.H10MO.D,30385,0.000608143506755,5.11408e-05 570 | HXC8_HUMAN.H10MO.D,22509,0.000470381622406,2.53022e-05 571 | PAX2_HUMAN.H10MO.S,50810,0.00136066651431,2.29776e-05 572 | SRBP1_HUMAN.H10MO.B,73045,0.00140106970201,0.00019744 573 | MEIS3_HUMAN.H10MO.D,11602,0.000318319528546,-1.27554e-05 574 | PAX2_HUMAN.H10MO.D,20938,0.000520796719215,4.29153e-06 575 | FOXP3_HUMAN.H10MO.D,27449,0.00108163964162,5.98431e-05 576 | MSX2_HUMAN.H10MO.D,21278,0.000503290572167,2.90573e-05 577 | HXA2_HUMAN.H10MO.D,18584,0.000468856472285,3.8743e-06 578 | HTF4_HUMAN.H10MO.B,24471,0.000659470885119,-1.37985e-05 579 | ARI3A_HUMAN.H10MO.D,50598,0.00113297418622,0.000229567 580 | OLIG2_HUMAN.H10MO.D,28372,0.000452093930508,-2.07722e-05 581 | CR3L1_HUMAN.H10MO.D,34304,0.000795156652946,-2.89977e-05 582 | SP2_HUMAN.H10MO.C,68729,0.00082809766291,0.000307977 583 | PIT1_HUMAN.H10MO.C,28033,0.000686991951361,2.19345e-05 584 | TF7L1_HUMAN.H10MO.D,85809,0.00183903071158,0.000104964 585 | FOSL2_HUMAN.H10MO.A,10720,0.000243265237584,-1.10269e-05 586 | ZFHX3_HUMAN.H10MO.D,50719,0.00153247052923,0.000124991 587 | PO3F2_HUMAN.H10MO.D,40415,0.00108907559696,8.43406e-05 588 | PRD14_HUMAN.H10MO.C,12243,0.000266406772981,9.41753e-06 589 | KLF15_HUMAN.H10MO.D,13625,0.000324762516112,0.000170708 590 | SPIC_HUMAN.H10MO.D,61806,0.00155273205853,0.00033316 591 | ZFX_HUMAN.H10MO.C,81891,0.00186000088798,0.000151306 592 | MNT_HUMAN.H10MO.D,77619,0.00197383090958,0.000560611 593 | LHX3_HUMAN.H10MO.C,45285,0.00105832593102,6.90818e-05 594 | MTF1_HUMAN.H10MO.C,77755,0.00156838399689,0.000122249 595 | LHX2_HUMAN.H10MO.D,20679,0.000450331725721,1.63019e-05 596 | FOXH1_HUMAN.H10MO.A,26224,0.000718944364427,4.38392e-05 597 | ZEP2_HUMAN.H10MO.D,48403,0.00119036334182,-3.47495e-05 598 | SOX17_HUMAN.H10MO.D,28003,0.000803750415264,3.76403e-05 599 | FIGLA_HUMAN.H10MO.D,20315,0.00027306371117,-3.96371e-06 600 | AP2A_HUMAN.H10MO.C,28824,0.000497282285916,1.02222e-05 601 | FOXM1_HUMAN.H10MO.D,48720,0.00129166634918,0.000274628 602 | PO2F2_HUMAN.H10MO.D,29461,0.000569485743217,4.68493e-05 603 | ETV1_HUMAN.H10MO.B,62715,0.00126624036779,0.000172883 604 | GSC2_HUMAN.H10MO.D,84164,0.00200884867436,0.000485837 605 | HXB2_HUMAN.H10MO.D,5657,0.000127020116055,-6.97374e-06 606 | RUNX2_HUMAN.H10MO.B,31992,0.000982168306168,5.52833e-05 607 | FOXO4_HUMAN.H10MO.C,60215,0.00122152161865,0.000313252 608 | E2F8_HUMAN.H10MO.D,36588,0.00110710424824,4.52995e-06 609 | BHA15_HUMAN.H10MO.D,38370,0.000505200209025,-1.93417e-05 610 | GCM2_HUMAN.H10MO.D,51401,0.000962277855886,-1.07288e-06 611 | ESR2_HUMAN.H10MO.A,28861,0.000779052487469,2.44975e-05 612 | NR2F6_HUMAN.H10MO.D,58166,0.00131828290936,5.5939e-05 613 | HSF2_HUMAN.H10MO.A,7803,9.34101218828e-05,-2.17557e-05 614 | LBX2_HUMAN.H10MO.D,15165,0.0001074740554,2.65241e-06 615 | TEAD4_HUMAN.H10MO.A,17866,0.000329252582377,4.70877e-06 616 | PROX1_HUMAN.H10MO.D,19220,0.000615850939399,0.000177592 617 | HEY2_HUMAN.H10MO.D,67887,0.00126021647044,0.000177473 618 | ESR2_HUMAN.H10MO.S,24060,0.000552224427427,-3.2872e-05 619 | AHR_HUMAN.H10MO.B,51426,0.00138095653836,1.00434e-05 620 | KLF1_HUMAN.H10MO.C,68036,0.00172423594267,0.000241071 621 | ZN639_HUMAN.H10MO.D,59385,0.00111830974478,0.000316262 622 | ALX3_HUMAN.H10MO.D,31428,0.000783796892928,4.99189e-05 623 | PLAL1_HUMAN.H10MO.D,68098,0.00170034702461,7.02143e-05 624 | ZEP1_HUMAN.H10MO.D,18351,0.000575367966075,-2.12789e-05 625 | BATF_HUMAN.H10MO.S,8089,0.000253811775726,-1.73151e-05 626 | DLX2_HUMAN.H10MO.D,21754,0.000602877018096,6.31809e-06 627 | FOXB1_HUMAN.H10MO.D,27210,0.000408440118936,8.35061e-05 628 | NF2L2_HUMAN.H10MO.D,13405,0.000432395692458,-9.83477e-07 629 | PRGR_HUMAN.H10MO.S,9432,0.000307097519276,-2.92063e-06 630 | STAT3_HUMAN.H10MO.A,16456,0.000356543920425,-1.43945e-05 631 | HXD12_HUMAN.H10MO.D,24990,0.000735722261249,1.72257e-05 632 | OVOL1_HUMAN.H10MO.C,25345,0.000622431505298,-2.26498e-06 633 | BATF_HUMAN.H10MO.A,9827,0.000173269602663,-3.75509e-06 634 | MESP1_HUMAN.H10MO.D,29038,0.000684707094729,-1.80304e-05 635 | KLF13_HUMAN.H10MO.D,87431,0.00180899063574,0.00030309 636 | ZBTB4_HUMAN.H10MO.D,47394,0.000942361492651,0.000121266 637 | HSFY1_HUMAN.H10MO.D,21407,0.000361279409346,1.7941e-05 638 | TFEB_HUMAN.H10MO.C,25223,0.000400725423665,-2.33054e-05 639 | HXA13_HUMAN.H10MO.C,38224,0.00118774578501,5.8502e-05 640 | ZBTB4_HUMAN.H10MO.S,77993,0.00134734410976,0.000206769 641 | -------------------------------------------------------------------------------- /feature_importances/SPEID/from_HOCOMOCO_motifs/NHEK_enhancers_feature_importance.csv: -------------------------------------------------------------------------------- 1 | Motif Name,Motif Count,AUPR Difference,MS Difference 2 | ETS1_HUMAN.H10MO.C,11618,0.000337757728758,-7.12276e-06 3 | HNF6_HUMAN.H10MO.C,15189,0.000362668551712,8.89897e-05 4 | FOXD3_HUMAN.H10MO.D,35332,0.000648370437891,0.000463188 5 | ZBTB6_HUMAN.H10MO.D,10153,0.00040094494805,3.82364e-05 6 | NR1H2_HUMAN.H10MO.D,13037,0.000302290638179,2.22325e-05 7 | ETV2_HUMAN.H10MO.D,29581,0.000437433608461,0.000265777 8 | ETV3_HUMAN.H10MO.D,19800,0.00034919698475,5.72205e-05 9 | RFX5_HUMAN.H10MO.A,18309,0.000499805647047,8.42512e-05 10 | SCRT2_HUMAN.H10MO.D,16583,0.000261217371425,-2.69711e-05 11 | ELF2_HUMAN.H10MO.C,79743,0.00135901097468,0.000599056 12 | MLXPL_HUMAN.H10MO.D,31575,0.000820709326684,2.02656e-06 13 | PAX6_HUMAN.H10MO.D,16739,0.000636011316544,-1.76728e-05 14 | MAX_HUMAN.H10MO.A,34334,0.00081748283542,2.16961e-05 15 | KLF6_HUMAN.H10MO.D,47881,0.000962838965619,0.000445783 16 | HNF1B_HUMAN.H10MO.B,16654,0.000469650213374,4.81308e-05 17 | PBX2_HUMAN.H10MO.C,13223,0.000424950176035,2.16365e-05 18 | RXRA_HUMAN.H10MO.C,27226,0.000606791842602,0.000167549 19 | GLIS2_HUMAN.H10MO.D,81192,0.00237108415673,0.000998855 20 | NFKB2_HUMAN.H10MO.D,15288,0.000351946129494,-1.69277e-05 21 | HSF4_HUMAN.H10MO.D,5010,8.390065093e-05,-1.21295e-05 22 | ERR2_HUMAN.H10MO.A,8007,0.000225202579839,1.40071e-06 23 | ESX1_HUMAN.H10MO.D,22541,0.000256426439668,6.55651e-05 24 | AP2B_HUMAN.H10MO.B,59965,0.00108409168605,0.000388265 25 | SPI1_HUMAN.H10MO.A,38053,0.000495507244843,0.000202924 26 | MYBA_HUMAN.H10MO.D,23459,0.000646997859623,8.07047e-05 27 | GFI1B_HUMAN.H10MO.C,10287,0.000421930277021,-1.07288e-06 28 | PO6F1_HUMAN.H10MO.D,19290,0.000390286628877,2.30968e-05 29 | ERR1_HUMAN.H10MO.D,38909,0.00108618109262,0.000143915 30 | PLAG1_HUMAN.H10MO.S,75771,0.00147430604938,0.000636607 31 | MAFA_HUMAN.H10MO.D,14251,0.000293114186508,4.50909e-05 32 | ELF1_HUMAN.H10MO.A,27672,0.000549349187312,0.000163019 33 | SPDEF_HUMAN.H10MO.D,16930,0.000418485019601,1.93715e-05 34 | GRHL1_HUMAN.H10MO.D,18413,0.000465321886227,3.06666e-05 35 | SCRT1_HUMAN.H10MO.D,12649,0.000318592254321,-6.94394e-06 36 | FOXJ2_HUMAN.H10MO.C,36719,0.000670331162335,0.000421405 37 | ASCL2_HUMAN.H10MO.D,72850,0.00188906924229,0.000965655 38 | XBP1_HUMAN.H10MO.C,29324,0.000475196066954,4.90844e-05 39 | FOXF2_HUMAN.H10MO.D,20013,0.000674520258433,0.00019747 40 | NFIC_HUMAN.H10MO.A,15800,0.00026323990479,3.60906e-05 41 | MYC_HUMAN.H10MO.A,41835,0.000638620335498,2.39909e-05 42 | STA5B_HUMAN.H10MO.C,4965,3.37136457297e-05,1.41859e-05 43 | ATF2_HUMAN.H10MO.B,14402,0.000281295891421,-1.07884e-05 44 | RELB_HUMAN.H10MO.C,16777,0.000390404742414,4.32134e-06 45 | HNF4G_HUMAN.H10MO.C,18042,0.000458856469321,3.36766e-05 46 | IRF4_HUMAN.H10MO.C,82876,0.00143114499764,0.000666201 47 | MEOX2_HUMAN.H10MO.D,20202,0.000497438219363,5.14686e-05 48 | DBP_HUMAN.H10MO.B,14263,0.000458769895932,1.2368e-05 49 | ANDR_HUMAN.H10MO.A,21458,0.000299285641481,0.000131547 50 | P63_HUMAN.H10MO.A,13464,0.000226111147076,1.20401e-05 51 | AIRE_HUMAN.H10MO.C,18955,0.000290026029757,5.27203e-05 52 | NFYA_HUMAN.H10MO.A,30075,0.000515466108974,8.50856e-05 53 | PO5F1_HUMAN.H10MO.A,19844,0.000577204240433,8.28803e-05 54 | SOX8_HUMAN.H10MO.D,12418,0.000320999391334,1.51992e-05 55 | THA_HUMAN.H10MO.C,14924,0.000193452786753,2.11596e-05 56 | INSM1_HUMAN.H10MO.C,47716,0.00117169046751,0.00024429 57 | CDX2_HUMAN.H10MO.C,13964,0.000259614409208,4.53889e-05 58 | ETS2_HUMAN.H10MO.C,35194,0.000755284565972,0.000204951 59 | ZN713_HUMAN.H10MO.D,56627,0.00140936883872,0.000769436 60 | MYF6_HUMAN.H10MO.C,9705,0.000124050294309,1.42455e-05 61 | RX_HUMAN.H10MO.D,19690,0.000216328429479,4.05908e-05 62 | GATA3_HUMAN.H10MO.C,11694,0.000304502540704,7.45058e-05 63 | ZN148_HUMAN.H10MO.D,54304,0.000781427298786,0.000427485 64 | PO3F4_HUMAN.H10MO.D,21467,0.000524192636335,7.34925e-05 65 | ZN784_HUMAN.H10MO.D,21947,0.000521665293414,0.000166625 66 | ZN524_HUMAN.H10MO.D,46879,0.00116427118191,0.000415713 67 | STAT6_HUMAN.H10MO.C,7070,9.20725720986e-05,1.7345e-05 68 | PAX1_HUMAN.H10MO.D,19551,0.000580807849273,-7.30157e-06 69 | PRRX1_HUMAN.H10MO.D,4215,-2.13987265286e-05,-3.12924e-06 70 | ZIC4_HUMAN.H10MO.D,54746,0.00114242462865,0.000379533 71 | CUX1_HUMAN.H10MO.C,15667,0.000439181291912,-9.53674e-07 72 | BARH1_HUMAN.H10MO.D,18226,0.000433979350622,4.69685e-05 73 | NKX21_HUMAN.H10MO.D,7797,0.000260375153958,-8.40425e-06 74 | HXD4_HUMAN.H10MO.D,14034,0.000456005271815,1.88947e-05 75 | PRRX2_HUMAN.H10MO.C,16147,0.000499062040598,1.71959e-05 76 | PO4F3_HUMAN.H10MO.D,41613,0.000500290584038,0.000199407 77 | HXB6_HUMAN.H10MO.D,20832,0.000736705666756,3.40939e-05 78 | NDF2_HUMAN.H10MO.D,32506,0.00109505397013,7.9006e-05 79 | HXD3_HUMAN.H10MO.D,15082,0.000401325091246,-9.89437e-06 80 | SRF_HUMAN.H10MO.A,17724,0.000468873371349,3.09646e-05 81 | ITF2_HUMAN.H10MO.B,14436,0.000190577761176,-2.75373e-05 82 | MYCN_HUMAN.H10MO.B,40332,0.000708685659734,1.40369e-05 83 | TFE2_HUMAN.H10MO.C,9050,0.000215544369237,-1.94311e-05 84 | FLI1_HUMAN.H10MO.A,30992,0.000662215231373,0.000140607 85 | PURA_HUMAN.H10MO.D,78381,0.000787729221641,0.000679553 86 | ETV7_HUMAN.H10MO.D,13449,0.000150713167034,8.49366e-06 87 | NFIA_HUMAN.H10MO.C,18979,0.000568549832362,7.17044e-05 88 | LHX8_HUMAN.H10MO.D,11650,0.000312533224033,1.75834e-06 89 | COE1_HUMAN.H10MO.A,15502,6.6553120874e-05,2.13683e-05 90 | KLF8_HUMAN.H10MO.C,37918,0.000786542160467,0.000166744 91 | PTF1A_HUMAN.H10MO.C,17000,0.000344680509041,-4.88758e-06 92 | STA5A_HUMAN.H10MO.B,5166,0.000115626858776,1.09673e-05 93 | ZSC16_HUMAN.H10MO.D,11542,0.000227157015349,5.0962e-06 94 | TBX21_HUMAN.H10MO.D,15489,0.000253663027486,4.18127e-05 95 | GATA5_HUMAN.H10MO.D,9096,0.000274978954637,2.53916e-05 96 | HES1_HUMAN.H10MO.D,37683,0.000910681353185,0.000120491 97 | TFCP2_HUMAN.H10MO.D,19034,0.000322749843434,2.77162e-05 98 | TGIF1_HUMAN.H10MO.S,7070,0.000264538005818,2.58684e-05 99 | BACH1_HUMAN.H10MO.A,12357,0.000446767718446,1.01924e-05 100 | GSX1_HUMAN.H10MO.D,19361,0.000394894490146,7.52807e-05 101 | TGIF1_HUMAN.H10MO.D,13963,0.00039012302372,2.31564e-05 102 | SOX10_HUMAN.H10MO.D,13762,0.000367724471228,0.000123233 103 | RARB_HUMAN.H10MO.D,18585,0.000300669289746,2.38419e-06 104 | FOXC2_HUMAN.H10MO.D,10236,0.000255046250567,4.32432e-05 105 | NFAC4_HUMAN.H10MO.C,21419,0.0006518510544,9.09567e-05 106 | TBX5_HUMAN.H10MO.D,10424,0.000210307880546,2.55406e-05 107 | UNC4_HUMAN.H10MO.D,25121,0.000517632259483,8.82447e-05 108 | MAFG_HUMAN.H10MO.S,6609,0.000229071024561,-1.54972e-06 109 | DRGX_HUMAN.H10MO.D,19076,0.00066010558874,6.88434e-05 110 | BHE41_HUMAN.H10MO.D,25111,0.000616269751956,7.45654e-05 111 | GLIS3_HUMAN.H10MO.D,74889,0.00192548969186,0.000734538 112 | GBX2_HUMAN.H10MO.D,22380,0.000517194462941,6.20484e-05 113 | ELF3_HUMAN.H10MO.D,18249,0.000115246372592,9.42945e-05 114 | TF2LX_HUMAN.H10MO.D,7734,0.0001288735468,2.23219e-05 115 | FOXD1_HUMAN.H10MO.D,12442,0.000353637187532,3.44217e-05 116 | CREB1_HUMAN.H10MO.A,20090,0.000331298017144,3.33488e-05 117 | EVX2_HUMAN.H10MO.A,23361,0.000427146925262,0.000163227 118 | BSH_HUMAN.H10MO.D,21986,0.000631927261086,2.81036e-05 119 | ZIC3_HUMAN.H10MO.C,42975,0.000791663736004,0.000328362 120 | ISL2_HUMAN.H10MO.D,9958,0.000165639972025,-1.42753e-05 121 | ZIC2_HUMAN.H10MO.C,79809,0.0021518083813,0.000880629 122 | EGR2_HUMAN.H10MO.C,53702,0.00114888125998,0.000656724 123 | FOXQ1_HUMAN.H10MO.C,29896,0.000970146278428,0.000282168 124 | ERR3_HUMAN.H10MO.B,6467,0.000179341694529,-2.68221e-07 125 | GMEB2_HUMAN.H10MO.D,15462,0.000329474910214,9.68575e-06 126 | JDP2_HUMAN.H10MO.D,17157,0.000249009960485,3.18885e-06 127 | SMAD3_HUMAN.H10MO.C,12737,7.26617219037e-05,5.62668e-05 128 | ETV5_HUMAN.H10MO.D,11545,0.000282786146466,2.12789e-05 129 | NKX28_HUMAN.H10MO.C,6516,0.000256867508778,-5.45382e-06 130 | OLIG3_HUMAN.H10MO.D,18835,0.000263121408583,1.59144e-05 131 | LMX1A_HUMAN.H10MO.D,41992,0.00117010675725,0.000277758 132 | HNF1A_HUMAN.H10MO.A,17162,0.000500244652658,3.62992e-05 133 | NANOG_HUMAN.H10MO.S,17496,0.000750152132487,1.98483e-05 134 | CREM_HUMAN.H10MO.C,19787,0.000488247509842,2.49743e-05 135 | MAFG_HUMAN.H10MO.C,5911,0.000139998273004,4.47035e-07 136 | PRDM1_HUMAN.H10MO.C,30416,0.000324862750913,0.000179172 137 | NANOG_HUMAN.H10MO.A,14192,0.000505831110714,5.57303e-05 138 | ARI5B_HUMAN.H10MO.C,15558,0.000564316624771,8.34763e-05 139 | MEIS2_HUMAN.H10MO.B,10192,0.000384609159233,-6.64592e-06 140 | ZN384_HUMAN.H10MO.C,26909,0.00055488935835,0.000149906 141 | PITX2_HUMAN.H10MO.D,17337,0.000412739884128,1.55866e-05 142 | ATF1_HUMAN.H10MO.B,15227,0.000300407987877,5.87106e-06 143 | STAT2_HUMAN.H10MO.B,39339,0.000942139146806,0.000340223 144 | ELK3_HUMAN.H10MO.D,22387,0.000556616168206,-9.59635e-06 145 | NOTO_HUMAN.H10MO.D,14158,0.000398300997768,2.31862e-05 146 | SP4_HUMAN.H10MO.D,96211,0.000983992589261,0.000792801 147 | ZBT18_HUMAN.H10MO.D,10645,0.000118895944246,-2.2471e-05 148 | ELF5_HUMAN.H10MO.D,13636,0.000284124565731,2.97129e-05 149 | NKX61_HUMAN.H10MO.D,31258,0.000710914532758,0.000172943 150 | PKNX1_HUMAN.H10MO.D,11710,0.00011145625839,1.35303e-05 151 | MSX1_HUMAN.H10MO.D,19862,0.000552116792181,6.80387e-05 152 | DLX4_HUMAN.H10MO.D,26571,0.000527385664415,8.4132e-05 153 | REST_HUMAN.H10MO.A,20479,0.000436132155933,8.05259e-05 154 | RARA_HUMAN.H10MO.S,4282,5.61209177691e-05,1.23382e-05 155 | FOSL1_HUMAN.H10MO.A,7541,0.000310297293349,-5.06639e-06 156 | FUBP1_HUMAN.H10MO.D,60912,0.00145050457893,0.000959128 157 | REL_HUMAN.H10MO.C,20291,0.000434178206694,2.27988e-05 158 | PO4F1_HUMAN.H10MO.D,38929,0.000551108637962,0.00017637 159 | PHX2A_HUMAN.H10MO.D,18637,0.000569392980258,6.57439e-05 160 | PO2F1_HUMAN.H10MO.B,16724,0.000344815494084,1.79112e-05 161 | HXB8_HUMAN.H10MO.C,25457,0.000745355857482,6.22272e-05 162 | RFX2_HUMAN.H10MO.C,29296,0.000705525085981,0.000108689 163 | VAX1_HUMAN.H10MO.D,19782,0.000331625074871,4.83692e-05 164 | EOMES_HUMAN.H10MO.D,14409,0.000489269484352,5.25713e-05 165 | FOS_HUMAN.H10MO.A,8509,0.000227132216631,-1.34408e-05 166 | ISL1_HUMAN.H10MO.D,13784,0.000507048056528,1.13845e-05 167 | LEF1_HUMAN.H10MO.C,8257,0.000282644339418,2.83122e-06 168 | FOXJ3_HUMAN.H10MO.S,59877,0.0011724113644,0.000863791 169 | SRBP2_HUMAN.H10MO.B,94327,0.0020889094744,0.00100648 170 | ATF6A_HUMAN.H10MO.B,50345,0.00091209784781,0.000149608 171 | TFDP1_HUMAN.H10MO.S,33687,0.000838625692458,9.11951e-05 172 | ZN423_HUMAN.H10MO.D,24099,0.000618013761466,4.01139e-05 173 | NR4A1_HUMAN.H10MO.C,12110,0.000457131567211,3.36766e-05 174 | NR2C2_HUMAN.H10MO.A,28784,0.000774683480277,0.00013414 175 | PPARG_HUMAN.H10MO.A,37157,0.000934172486453,0.000251651 176 | TFDP1_HUMAN.H10MO.D,52096,0.000819970275773,0.000690341 177 | RFX4_HUMAN.H10MO.D,13623,0.000273135027299,4.30942e-05 178 | CEBPG_HUMAN.H10MO.C,19699,0.000375251265825,-5.96046e-07 179 | NFAC1_HUMAN.H10MO.S,12691,0.000314110550804,-1.08778e-05 180 | PPARG_HUMAN.H10MO.S,12154,0.000393843478748,1.56462e-05 181 | SOX2_HUMAN.H10MO.B,16752,0.000420297421814,6.92606e-05 182 | TBX1_HUMAN.H10MO.D,61630,0.000628488116491,0.000557661 183 | SOX7_HUMAN.H10MO.D,14433,0.000425548139081,8.12709e-05 184 | GATA1_HUMAN.H10MO.S,8918,0.000375613293673,2.94149e-05 185 | MYOD1_HUMAN.H10MO.C,12556,0.000306630198713,-4.17233e-07 186 | WT1_HUMAN.H10MO.D,54505,0.000990823920553,0.000780761 187 | BHE22_HUMAN.H10MO.D,26122,0.000296549477672,4.40776e-05 188 | GATA1_HUMAN.H10MO.A,13081,0.000227818788242,3.82364e-05 189 | ERG_HUMAN.H10MO.B,22624,0.000230187033318,-1.2815e-06 190 | E2F7_HUMAN.H10MO.D,31227,0.000539083249934,0.000114948 191 | HAND1_HUMAN.H10MO.D,7804,0.000190317899236,2.14875e-05 192 | NKX31_HUMAN.H10MO.C,32287,0.000619637943165,0.000229508 193 | HXD8_HUMAN.H10MO.D,14308,0.000458306483459,-8.58307e-06 194 | ELK1_HUMAN.H10MO.A,15961,0.000214335509708,-1.85668e-05 195 | FOSB_HUMAN.H10MO.C,8129,0.000103420006931,2.5034e-06 196 | TBX2_HUMAN.H10MO.D,9256,0.000354878376841,3.09944e-06 197 | FOXO3_HUMAN.H10MO.B,28671,0.000623037645585,0.000310987 198 | RUNX3_HUMAN.H10MO.C,17792,0.00075202024439,8.36551e-05 199 | NKX23_HUMAN.H10MO.D,11596,0.000401595173065,-9.0301e-06 200 | NKX62_HUMAN.H10MO.D,18267,0.000413068541104,4.05312e-06 201 | ESR1_HUMAN.H10MO.S,17894,0.000307951299902,2.76268e-05 202 | MAZ_HUMAN.H10MO.A,58709,0.000892004463191,0.00054121 203 | HXA7_HUMAN.H10MO.D,6830,0.000216299978829,7.92742e-06 204 | CRX_HUMAN.H10MO.C,9788,0.000428165344363,1.21891e-05 205 | SHOX_HUMAN.H10MO.D,26854,0.000577070430267,0.000100702 206 | NOBOX_HUMAN.H10MO.C,11951,0.000207741515186,-1.92523e-05 207 | MZF1_HUMAN.H10MO.D,84401,0.00198703118556,0.000823706 208 | STAT1_HUMAN.H10MO.A,13618,0.000251251404473,9.32813e-06 209 | CTCFL_HUMAN.H10MO.A,32129,0.000611709878089,0.00013417 210 | CEBPD_HUMAN.H10MO.B,21431,0.000302508681398,2.88188e-05 211 | TF7L2_HUMAN.H10MO.A,14023,0.000346044262131,6.71446e-05 212 | ID4_HUMAN.H10MO.D,15622,9.05020162639e-05,-2.22027e-05 213 | PAX3_HUMAN.H10MO.D,18735,0.000306193139509,4.92632e-05 214 | STAT1_HUMAN.H10MO.S,9402,0.000254097572693,-1.13547e-05 215 | PAX7_HUMAN.H10MO.D,19942,0.000538411707885,5.86808e-05 216 | NKX22_HUMAN.H10MO.D,7729,6.911827671e-05,1.3411e-05 217 | RXRG_HUMAN.H10MO.B,15020,0.000203659134538,3.60012e-05 218 | TBX20_HUMAN.H10MO.D,39357,0.000743837894186,0.000346333 219 | MUSC_HUMAN.H10MO.D,21410,0.000667297000839,4.64916e-05 220 | BC11A_HUMAN.H10MO.C,75831,0.00165489505356,0.000698775 221 | VSX2_HUMAN.H10MO.D,7541,0.000180428073786,-2.46465e-05 222 | NFAT5_HUMAN.H10MO.D,12473,0.00023059220718,2.3216e-05 223 | BCL6_HUMAN.H10MO.C,4555,0.000129325832376,-5.93066e-06 224 | ZIC1_HUMAN.H10MO.B,84621,0.00176644576649,0.00089702 225 | MEF2D_HUMAN.H10MO.C,11077,0.000126210811244,6.95586e-05 226 | SP3_HUMAN.H10MO.B,79014,0.000860794236952,0.000611752 227 | NR1H4_HUMAN.H10MO.C,17516,0.000431787411188,7.26879e-05 228 | E2F4_HUMAN.H10MO.A,42259,0.00124332455826,0.000143737 229 | E2F5_HUMAN.H10MO.B,35214,0.000624778746484,0.000118434 230 | HLF_HUMAN.H10MO.C,20727,0.000558186445947,1.63317e-05 231 | TBX4_HUMAN.H10MO.D,13517,0.000167007558796,2.86102e-06 232 | HXD13_HUMAN.H10MO.D,10269,0.000116998339242,1.43647e-05 233 | HME1_HUMAN.H10MO.D,21817,0.000445956515564,6.41942e-05 234 | TYY1_HUMAN.H10MO.A,21961,0.00037178658463,3.8445e-05 235 | FOXF1_HUMAN.H10MO.D,39559,0.00082247710864,0.00042209 236 | EMX1_HUMAN.H10MO.D,17666,0.0001660340401,3.1352e-05 237 | ZN232_HUMAN.H10MO.D,13419,0.000301844518862,2.0653e-05 238 | ZSCA4_HUMAN.H10MO.D,22168,0.000760959994175,0.000117719 239 | SOX13_HUMAN.H10MO.D,25683,0.000602501596828,0.000160277 240 | NKX32_HUMAN.H10MO.C,15927,0.00049562054168,6.4373e-06 241 | CREB3_HUMAN.H10MO.D,36187,0.000670660052495,8.98242e-05 242 | PO2F3_HUMAN.H10MO.D,14100,0.000452224031717,2.02656e-06 243 | HMX2_HUMAN.H10MO.D,15102,0.0002557207637,5.25415e-05 244 | SNAI1_HUMAN.H10MO.C,16350,0.000290077320269,-6.10948e-06 245 | HXC6_HUMAN.H10MO.D,31785,0.000619153829105,0.00039506 246 | P63_HUMAN.H10MO.S,11552,0.000259622614671,-4.23193e-06 247 | VAX2_HUMAN.H10MO.D,12517,0.000270125315268,-7.36117e-06 248 | CEBPB_HUMAN.H10MO.A,17139,0.000160304416811,5.57303e-06 249 | TLX1_HUMAN.H10MO.S,17561,0.000537730540903,4.12464e-05 250 | BATF3_HUMAN.H10MO.D,21695,0.00070217422707,2.78056e-05 251 | HXB1_HUMAN.H10MO.D,18238,0.000352403976778,1.44839e-05 252 | PDX1_HUMAN.H10MO.C,7572,0.000219500875119,-4.38094e-06 253 | TF65_HUMAN.H10MO.A,18116,0.000439302966553,2.50936e-05 254 | LHX6_HUMAN.H10MO.D,15754,0.000277771119121,2.61068e-05 255 | E4F1_HUMAN.H10MO.D,28410,0.000566239043984,7.80225e-05 256 | BHE23_HUMAN.H10MO.D,30588,0.000638056015214,0.000144005 257 | SOX3_HUMAN.H10MO.D,15809,0.000379576165418,7.49528e-05 258 | PEBB_HUMAN.H10MO.C,18507,0.000663694194812,6.21676e-05 259 | TLX1_HUMAN.H10MO.D,13081,0.000162078767777,2.35438e-05 260 | NFYB_HUMAN.H10MO.A,24938,0.000630848715443,4.67598e-05 261 | ALX4_HUMAN.H10MO.D,21260,0.000664000082202,8.26716e-05 262 | MAFK_HUMAN.H10MO.A,5777,0.000142097278209,-9.56655e-06 263 | CENPB_HUMAN.H10MO.D,37031,0.00050487983577,0.000215292 264 | HME2_HUMAN.H10MO.D,12865,0.000190437809735,-1.07586e-05 265 | BARX1_HUMAN.H10MO.D,21772,0.000330963862392,0.000105798 266 | HXC11_HUMAN.H10MO.D,16672,0.000334985310364,3.8892e-05 267 | HXB3_HUMAN.H10MO.D,19580,0.000292975120833,9.85563e-05 268 | ATF7_HUMAN.H10MO.D,16700,0.000302600445283,8.70228e-06 269 | OLIG1_HUMAN.H10MO.D,20470,0.000282080345115,2.59578e-05 270 | CR3L2_HUMAN.H10MO.D,47090,0.000872819966262,0.000175893 271 | JUND_HUMAN.H10MO.A,7975,7.98310053433e-05,-3.06964e-06 272 | CDC5L_HUMAN.H10MO.D,22384,0.000623042793756,6.73532e-05 273 | MEOX1_HUMAN.H10MO.D,16668,0.000432468531526,1.87457e-05 274 | EPAS1_HUMAN.H10MO.D,51808,0.000654532541787,0.000425071 275 | IRF3_HUMAN.H10MO.C,61774,0.00134770042541,0.000465631 276 | KAISO_HUMAN.H10MO.A,43541,0.000827123331799,0.000403672 277 | RUNX1_HUMAN.H10MO.A,16510,0.000566111710802,3.82662e-05 278 | PO3F3_HUMAN.H10MO.D,43851,0.000841502001207,0.000400901 279 | IRX2_HUMAN.H10MO.D,87700,0.00163191602518,0.000884712 280 | KAISO_HUMAN.H10MO.S,10802,0.000259354556766,5.58496e-05 281 | MYB_HUMAN.H10MO.C,17838,0.000403004983793,3.41833e-05 282 | GFI1_HUMAN.H10MO.C,9336,0.000160067562617,1.85668e-05 283 | ETV4_HUMAN.H10MO.B,9425,0.000295497398674,5.72205e-06 284 | NR5A2_HUMAN.H10MO.C,7116,0.000140324443274,2.26498e-06 285 | HBP1_HUMAN.H10MO.D,11730,0.000136577539821,1.78814e-06 286 | ARX_HUMAN.H10MO.D,17852,0.000581565813369,6.44326e-05 287 | EMX2_HUMAN.H10MO.D,18838,0.000362477255584,4.30644e-05 288 | ARNT2_HUMAN.H10MO.D,79576,0.00120685270398,0.000785679 289 | ZN333_HUMAN.H10MO.D,26620,0.000593051154164,8.84831e-05 290 | NR2E1_HUMAN.H10MO.D,14337,0.000254536615622,5.22137e-05 291 | FOXA3_HUMAN.H10MO.C,30545,0.00078064578415,0.000306487 292 | BARX2_HUMAN.H10MO.D,20240,0.00056642412906,9.14931e-06 293 | TEF_HUMAN.H10MO.D,18863,0.000761115951485,6.49095e-05 294 | ONEC3_HUMAN.H10MO.D,15683,0.000459865644547,0.000112832 295 | ONEC2_HUMAN.H10MO.D,25636,0.000755309379124,0.000280112 296 | ZN282_HUMAN.H10MO.D,34403,0.00106036354295,0.000117987 297 | HXB13_HUMAN.H10MO.D,14656,0.000310378626577,5.31375e-05 298 | MNT_HUMAN.H10MO.D,84162,0.00116809214702,0.00111887 299 | MIXL1_HUMAN.H10MO.D,13520,0.000348278985613,2.30968e-05 300 | FOXP2_HUMAN.H10MO.A,16133,0.000310658259442,4.81308e-05 301 | TEAD3_HUMAN.H10MO.D,10000,0.000150216560491,4.99785e-05 302 | TWST1_HUMAN.H10MO.D,21061,0.000391041285792,1.57058e-05 303 | OTX1_HUMAN.H10MO.D,4963,0.000254418758759,-2.5928e-06 304 | NR6A1_HUMAN.H10MO.B,11445,0.00039679559315,-1.37389e-05 305 | HXD9_HUMAN.H10MO.D,17238,0.00059641302087,4.69387e-05 306 | BRAC_HUMAN.H10MO.D,9258,0.000189429738487,1.88947e-05 307 | ERF_HUMAN.H10MO.D,16286,0.000565597219505,6.09159e-05 308 | FOXA1_HUMAN.H10MO.A,15837,0.000479788573706,6.66976e-05 309 | FOXJ3_HUMAN.H10MO.A,68022,0.00143954464556,0.00102213 310 | MYOG_HUMAN.H10MO.D,8408,4.0181660395e-05,1.54972e-06 311 | BARH2_HUMAN.H10MO.D,18631,0.000574750714226,1.90437e-05 312 | MAFB_HUMAN.H10MO.D,8700,0.000202011097381,1.09076e-05 313 | GLI1_HUMAN.H10MO.C,32530,0.000706804565173,0.000228167 314 | NGN2_HUMAN.H10MO.D,25170,0.000616407822634,2.63155e-05 315 | PLAG1_HUMAN.H10MO.D,50500,0.000791878013147,0.000466615 316 | E2F1_HUMAN.H10MO.A,62332,0.00186519349783,0.000411004 317 | YBOX1_HUMAN.H10MO.D,23668,0.000490747266419,2.88486e-05 318 | IRF1_HUMAN.H10MO.A,56176,0.00120920957678,0.000504285 319 | P5F1B_HUMAN.H10MO.D,20004,0.000464224715592,6.51777e-05 320 | HMX1_HUMAN.H10MO.D,20115,0.000316103363579,0.000113159 321 | SOX1_HUMAN.H10MO.D,12534,0.000319821321521,6.06775e-05 322 | TBX15_HUMAN.H10MO.D,92407,0.000929303354928,0.00064382 323 | PPARD_HUMAN.H10MO.D,18427,0.000682501512257,4.42863e-05 324 | RFX3_HUMAN.H10MO.B,10486,0.000296635158887,9.83477e-06 325 | USF2_HUMAN.H10MO.A,41290,0.000772787839808,3.76105e-05 326 | RREB1_HUMAN.H10MO.D,86812,0.000892674637622,0.000595093 327 | MAFF_HUMAN.H10MO.A,5814,0.000167651982869,1.89841e-05 328 | FOXD2_HUMAN.H10MO.D,26473,0.00058830884077,0.00023669 329 | E2F3_HUMAN.H10MO.B,47730,0.000678586199029,0.000167042 330 | PO4F2_HUMAN.H10MO.D,16154,0.000446645716917,3.84748e-05 331 | THA_HUMAN.H10MO.S,10509,0.000126204281783,-3.3766e-05 332 | EGR4_HUMAN.H10MO.D,44765,0.00131731697103,0.000656813 333 | HIC2_HUMAN.H10MO.D,22157,0.000508626636429,3.83854e-05 334 | NFIA_HUMAN.H10MO.S,3256,6.3824011716e-05,-5.96046e-08 335 | STAT4_HUMAN.H10MO.D,7351,0.000257912614806,1.11759e-05 336 | TGIF2_HUMAN.H10MO.D,8627,0.000308567325914,-1.68979e-05 337 | MCR_HUMAN.H10MO.D,11849,0.000372092714394,1.84476e-05 338 | PITX3_HUMAN.H10MO.D,15233,0.000358000097557,3.74019e-05 339 | SHOX2_HUMAN.H10MO.D,26188,0.000509085979641,0.000107229 340 | NR1D1_HUMAN.H10MO.C,11603,0.000192648684584,1.3411e-05 341 | MEIS1_HUMAN.H10MO.C,8866,0.000326605325846,1.15633e-05 342 | ZN740_HUMAN.H10MO.D,83249,0.0011514939144,0.000634372 343 | PRDM4_HUMAN.H10MO.D,15994,0.00039317859323,5.22733e-05 344 | GSC_HUMAN.H10MO.D,16709,0.000444014590891,3.19481e-05 345 | DMBX1_HUMAN.H10MO.D,25177,0.000615106861684,0.000127167 346 | PO6F2_HUMAN.H10MO.D,7595,6.26078408441e-05,1.46031e-06 347 | SOX5_HUMAN.H10MO.C,30485,0.000742205777025,0.000227243 348 | NRL_HUMAN.H10MO.D,10489,0.0002739808394,3.93391e-06 349 | ATF3_HUMAN.H10MO.A,14604,0.000358685211362,-1.89841e-05 350 | SP1_HUMAN.H10MO.C,63512,0.00076039177787,0.000604689 351 | ETV6_HUMAN.H10MO.D,22236,0.000561536810915,0.000132143 352 | IRF8_HUMAN.H10MO.D,35521,0.00104223894712,0.000189722 353 | SP1_HUMAN.H10MO.S,63891,0.00110322244481,0.000728726 354 | SPZ1_HUMAN.H10MO.D,52559,0.00127561033603,0.00041455 355 | BPTF_HUMAN.H10MO.D,66442,0.00146962735406,0.00112423 356 | PHX2B_HUMAN.H10MO.D,18877,0.000494711321578,5.82635e-05 357 | HSF1_HUMAN.H10MO.A,8409,0.000119752848493,-5.96046e-07 358 | CXXC1_HUMAN.H10MO.D,21300,0.000719925783608,6.8754e-05 359 | MYBB_HUMAN.H10MO.D,13541,0.000413259268123,5.03063e-05 360 | GATA6_HUMAN.H10MO.B,10559,0.000399875182085,7.26283e-05 361 | FOXK1_HUMAN.H10MO.D,56805,0.00177445319435,0.000513852 362 | BMAL1_HUMAN.H10MO.C,32804,0.000812008411516,5.39422e-06 363 | GLI2_HUMAN.H10MO.B,22682,0.000570994882994,0.000109136 364 | DLX6_HUMAN.H10MO.D,28339,0.00066633015081,0.000117391 365 | MAFK_HUMAN.H10MO.S,5779,1.50570065578e-05,4.94719e-06 366 | HXD11_HUMAN.H10MO.D,23220,0.000598121379564,0.0001809 367 | TFE3_HUMAN.H10MO.C,24299,0.000523121712153,-2.39015e-05 368 | TYY2_HUMAN.H10MO.D,37559,0.000696588990392,0.000125676 369 | HIF1A_HUMAN.H10MO.A,51101,0.0010419546725,9.97782e-05 370 | NKX25_HUMAN.H10MO.C,4738,0.000180074662651,-1.64509e-05 371 | KLF14_HUMAN.H10MO.D,85305,0.00217302365157,0.000959128 372 | NDF1_HUMAN.H10MO.C,21485,0.000585022710846,7.7486e-06 373 | NR0B1_HUMAN.H10MO.D,68987,0.0017727919495,0.000577688 374 | LHX9_HUMAN.H10MO.D,24921,0.000470957918651,9.32813e-05 375 | HXA10_HUMAN.H10MO.C,20782,0.000588485993996,5.85616e-05 376 | NRF1_HUMAN.H10MO.A,48716,0.000896348137773,0.000306964 377 | NF2L1_HUMAN.H10MO.C,6609,0.000186361131912,-8.58307e-06 378 | ATOH1_HUMAN.H10MO.D,19139,0.000547227070882,-1.16825e-05 379 | ZBED1_HUMAN.H10MO.D,13587,0.000314057254202,-3.72529e-06 380 | GATA4_HUMAN.H10MO.B,8467,0.000282715932859,2.55406e-05 381 | SOX9_HUMAN.H10MO.B,26833,0.000725737540468,0.000115275 382 | LHX4_HUMAN.H10MO.D,20149,0.000265978132201,3.48091e-05 383 | NR4A2_HUMAN.H10MO.C,10773,0.000432266523482,2.18153e-05 384 | MNX1_HUMAN.H10MO.D,25139,0.000342522094499,0.000157267 385 | MGAP_HUMAN.H10MO.D,13434,0.000264604941713,1.20103e-05 386 | AP2D_HUMAN.H10MO.D,95408,0.00188907954324,0.000562221 387 | BCL6B_HUMAN.H10MO.D,5875,0.000138161925903,1.72853e-06 388 | HXA11_HUMAN.H10MO.D,21493,0.000591216843358,0.000133544 389 | SMRC1_HUMAN.H10MO.D,8505,4.37063353893e-05,1.42455e-05 390 | HMGA1_HUMAN.H10MO.D,15955,0.000219803981678,0.000120491 391 | JUNB_HUMAN.H10MO.C,9324,0.00019474614918,2.68221e-07 392 | DLX1_HUMAN.H10MO.D,21442,0.000787402130736,8.58605e-05 393 | FOXA2_HUMAN.H10MO.A,16431,0.000401091168282,7.05719e-05 394 | IRF5_HUMAN.H10MO.D,51744,0.00133514933772,0.000680953 395 | NFYC_HUMAN.H10MO.B,33971,0.000584891888119,1.53184e-05 396 | ARNT_HUMAN.H10MO.B,39144,0.000970381021325,4.50313e-05 397 | RORG_HUMAN.H10MO.C,13938,0.000331507955489,3.67463e-05 398 | RAX2_HUMAN.H10MO.D,21916,0.000554864972244,6.75917e-05 399 | COT1_HUMAN.H10MO.B,10012,0.00023309723793,-6.55651e-06 400 | THB_HUMAN.H10MO.S,11138,0.000292596793159,-2.92063e-05 401 | ZEB1_HUMAN.H10MO.B,15048,0.000342333861343,-4.76837e-07 402 | SMAD4_HUMAN.H10MO.C,13330,0.000463115848782,5.45382e-05 403 | THB_HUMAN.H10MO.C,18079,0.000390485444626,3.19779e-05 404 | KLF12_HUMAN.H10MO.D,50811,0.000956532831387,0.000359118 405 | SOX21_HUMAN.H10MO.D,13852,0.000270166845173,5.57601e-05 406 | E2F6_HUMAN.H10MO.C,53508,0.00139745493934,0.000591666 407 | MBD2_HUMAN.H10MO.B,81346,0.00135416260009,0.000414044 408 | HXD10_HUMAN.H10MO.D,22181,0.000558254153341,0.000257939 409 | P73_HUMAN.H10MO.S,17766,0.000345071070472,1.0401e-05 410 | SMAD2_HUMAN.H10MO.C,12131,0.000401532384236,3.08454e-05 411 | EGR1_HUMAN.H10MO.A,81722,0.00118200573084,0.00093326 412 | JUN_HUMAN.H10MO.A,7797,0.00018144463294,7.83801e-06 413 | CEBPE_HUMAN.H10MO.A,18215,0.00047923603687,1.47223e-05 414 | RARG_HUMAN.H10MO.C,26174,0.000739405926977,0.000115424 415 | IRX3_HUMAN.H10MO.D,25113,0.000550414243756,0.000109076 416 | EGR3_HUMAN.H10MO.D,73315,0.000783108394098,0.000445068 417 | P73_HUMAN.H10MO.A,15720,0.000346936674273,3.23355e-05 418 | TBX19_HUMAN.H10MO.D,14142,0.00030409657243,2.65539e-05 419 | DUXA_HUMAN.H10MO.D,10230,0.000228237491024,2.3216e-05 420 | FOXL1_HUMAN.H10MO.D,51624,0.000597342398596,0.000754118 421 | EGR1_HUMAN.H10MO.S,68431,0.00161499491748,0.000690371 422 | HLTF_HUMAN.H10MO.D,14452,0.000244422898631,1.42455e-05 423 | RARG_HUMAN.H10MO.S,4282,0.000156747327386,-5.96046e-08 424 | COT2_HUMAN.H10MO.A,19013,0.000701871477379,5.69522e-05 425 | UBIP1_HUMAN.H10MO.D,5018,-2.02190644151e-05,1.56462e-05 426 | MAF_HUMAN.H10MO.B,6766,0.000234418426447,-1.49012e-07 427 | HXC13_HUMAN.H10MO.D,15431,0.000265173100842,4.45843e-05 428 | DDIT3_HUMAN.H10MO.C,18775,0.000392390462456,3.15905e-05 429 | CPEB1_HUMAN.H10MO.D,66929,0.000832964555134,0.000978947 430 | COT2_HUMAN.H10MO.S,23818,0.000749427374174,4.82202e-05 431 | PITX1_HUMAN.H10MO.D,18526,0.000449881828474,-3.96371e-06 432 | ENOA_HUMAN.H10MO.A,40972,0.000879061817179,4.76241e-05 433 | ZKSC3_HUMAN.H10MO.D,3635,0.000128977552961,1.35899e-05 434 | RXRB_HUMAN.H10MO.C,11648,0.000208893006245,-4.67598e-05 435 | NFE2_HUMAN.H10MO.B,8199,0.000224589661577,3.8445e-06 436 | TAL1_HUMAN.H10MO.A,11348,0.000304038923927,2.73585e-05 437 | SOX11_HUMAN.H10MO.D,13471,0.000278832680744,5.27203e-05 438 | GSX2_HUMAN.H10MO.D,23506,0.000699770227022,0.00012055 439 | TAL1_HUMAN.H10MO.S,16199,0.000323928143343,7.45058e-07 440 | BRCA1_HUMAN.H10MO.D,24346,0.000695401083631,0.000109047 441 | FEV_HUMAN.H10MO.C,13746,0.000282250138466,-5.126e-06 442 | VSX1_HUMAN.H10MO.D,21809,0.000415419147193,6.47604e-05 443 | PBX3_HUMAN.H10MO.B,13547,0.000379837622139,1.06394e-05 444 | MLX_HUMAN.H10MO.D,29591,0.000648142145752,-1.82688e-05 445 | IRF7_HUMAN.H10MO.C,28772,0.000608581745309,0.00014776 446 | PO3F1_HUMAN.H10MO.C,16947,0.000331103067371,5.30779e-05 447 | SRY_HUMAN.H10MO.B,45905,0.00164201824285,0.000569999 448 | HMX3_HUMAN.H10MO.D,14793,0.000448494386579,1.38879e-05 449 | SOX4_HUMAN.H10MO.C,24622,0.000884973293151,0.000169218 450 | CDX1_HUMAN.H10MO.C,18498,0.000371341545382,4.12166e-05 451 | RARA_HUMAN.H10MO.C,27699,0.000503497802162,0.000171661 452 | STF1_HUMAN.H10MO.B,9292,0.00024836322964,9.53674e-06 453 | HMBX1_HUMAN.H10MO.D,4360,0.000137953291052,-4.94719e-06 454 | ZN652_HUMAN.H10MO.D,23758,0.000490295891516,0.000106782 455 | MECP2_HUMAN.H10MO.C,21329,0.000264449322325,3.99947e-05 456 | VDR_HUMAN.H10MO.B,75853,0.00169146591547,0.00110576 457 | ZBT49_HUMAN.H10MO.D,13026,0.000306466316348,-2.2918e-05 458 | AP2C_HUMAN.H10MO.A,27951,0.000606079554522,9.11355e-05 459 | RFX1_HUMAN.H10MO.C,13348,0.000440019023077,4.19617e-05 460 | NFAC3_HUMAN.H10MO.B,17302,0.000151864044453,6.6638e-05 461 | NFAC1_HUMAN.H10MO.A,19289,0.000573432253268,2.2769e-05 462 | VDR_HUMAN.H10MO.S,15010,0.000375905176965,2.66135e-05 463 | NR2E3_HUMAN.H10MO.C,24608,0.000460953863207,0.000291556 464 | VENTX_HUMAN.H10MO.D,16153,0.000481947425593,4.96209e-05 465 | SOX15_HUMAN.H10MO.D,17445,0.000311167373946,6.88732e-05 466 | NR1I3_HUMAN.H10MO.C,11724,0.000274755934057,2.70903e-05 467 | CREB5_HUMAN.H10MO.D,15345,0.000267086455368,2.5928e-06 468 | RHXF1_HUMAN.H10MO.D,18239,0.000343037000499,1.03116e-05 469 | TCF7_HUMAN.H10MO.C,12022,0.000302860338678,3.73423e-05 470 | HOMEZ_HUMAN.H10MO.D,35810,0.000586315040022,0.000283659 471 | SUH_HUMAN.H10MO.C,15750,0.000408138051442,5.427e-05 472 | NR1I3_HUMAN.H10MO.S,8332,0.000260067770273,1.99676e-06 473 | GABP1_HUMAN.H10MO.C,28455,0.000561040874799,9.21488e-05 474 | FOXI1_HUMAN.H10MO.B,16322,0.000796363481422,5.51343e-06 475 | KLF4_HUMAN.H10MO.A,58390,0.00106124846058,0.000669509 476 | TBX3_HUMAN.H10MO.D,10796,0.000203331795822,3.46601e-05 477 | GLIS1_HUMAN.H10MO.D,67621,0.00197261519936,0.000704944 478 | DLX5_HUMAN.H10MO.D,21683,0.00058159501304,7.32839e-05 479 | DLX3_HUMAN.H10MO.C,11075,0.000215760898101,5.39422e-06 480 | TEAD1_HUMAN.H10MO.D,15715,0.000394178566856,1.27554e-05 481 | GATA2_HUMAN.H10MO.A,10815,0.000419335069207,4.99785e-05 482 | P53_HUMAN.H10MO.B,13991,0.000253014835034,-1.12951e-05 483 | CLOCK_HUMAN.H10MO.D,77742,0.00113851093659,0.000647485 484 | ALX1_HUMAN.H10MO.B,19350,0.000456080232492,6.52075e-05 485 | GLI3_HUMAN.H10MO.B,37007,0.000690618644745,0.000207931 486 | FOXG1_HUMAN.H10MO.D,51123,0.000965058645007,0.000884444 487 | HEY1_HUMAN.H10MO.D,39053,0.000738297043246,-1.28746e-05 488 | ZN219_HUMAN.H10MO.D,29133,0.000810023673998,0.000539839 489 | BHE40_HUMAN.H10MO.A,29209,0.000592287187277,-3.00407e-05 490 | MEF2B_HUMAN.H10MO.D,8261,0.000119537047809,2.48849e-05 491 | HMGA2_HUMAN.H10MO.D,26255,0.00055699950567,-1.87755e-06 492 | IRF2_HUMAN.H10MO.C,34237,0.000841898319736,0.000222057 493 | MEF2A_HUMAN.H10MO.A,10057,0.000241059019747,6.60419e-05 494 | ZBT7B_HUMAN.H10MO.D,25979,0.000760724164006,0.000100762 495 | HES5_HUMAN.H10MO.D,40916,0.000888321796502,-2.67029e-05 496 | HXA5_HUMAN.H10MO.D,22785,0.000338215695197,2.7597e-05 497 | ISX_HUMAN.H10MO.D,21691,0.000218898253399,7.75456e-05 498 | PAX5_HUMAN.H10MO.A,43806,0.000992336754067,0.000275671 499 | HES7_HUMAN.H10MO.D,34466,0.00082652749973,-2.80738e-05 500 | HNF4A_HUMAN.H10MO.A,14447,0.000244637076768,2.99215e-05 501 | NR2C1_HUMAN.H10MO.C,11736,0.000202948860375,9.0003e-06 502 | GBX1_HUMAN.H10MO.D,24912,0.000579716820469,7.86185e-05 503 | FOXC1_HUMAN.H10MO.C,18707,0.000753940065522,0.000151604 504 | MITF_HUMAN.H10MO.C,20973,0.000565619485801,-2.43187e-05 505 | PAX5_HUMAN.H10MO.S,3951,0.000115694722102,6.73532e-06 506 | COT1_HUMAN.H10MO.S,14228,0.000286027673229,6.67572e-06 507 | NFAC2_HUMAN.H10MO.B,21500,0.000443622812302,8.02875e-05 508 | PKNX2_HUMAN.H10MO.D,68498,0.00182543797685,0.000718474 509 | OTX2_HUMAN.H10MO.C,16046,0.000463223866298,2.6226e-06 510 | MEF2C_HUMAN.H10MO.C,14818,0.000371540597294,0.000168562 511 | NR1I2_HUMAN.H10MO.C,11564,0.000263962111674,8.91089e-06 512 | KLF16_HUMAN.H10MO.D,81389,0.000850726531659,0.000689983 513 | ZN589_HUMAN.H10MO.D,39484,0.000936874377108,0.000281215 514 | HXA1_HUMAN.H10MO.C,19656,0.000356799283501,1.37091e-05 515 | TBR1_HUMAN.H10MO.D,17303,0.000421352469125,4.20809e-05 516 | GABPA_HUMAN.H10MO.A,30839,0.000505414816945,0.000165343 517 | ELK4_HUMAN.H10MO.A,23759,0.000453750836183,6.68764e-05 518 | IRF9_HUMAN.H10MO.C,16861,0.000446841956527,6.49989e-05 519 | FOXO6_HUMAN.H10MO.D,18207,0.000540724283409,5.48363e-05 520 | DPRX_HUMAN.H10MO.D,16396,0.000355937650346,-5.72205e-06 521 | EVI1_HUMAN.H10MO.B,14329,0.000270069947183,0.000133216 522 | NFKB1_HUMAN.H10MO.B,19968,0.000537806676282,-6.10948e-06 523 | EHF_HUMAN.H10MO.S,19298,0.000498270034506,6.935e-05 524 | TFAP4_HUMAN.H10MO.C,14685,0.000234037451666,3.67463e-05 525 | ZKSC1_HUMAN.H10MO.C,21687,0.000397018127921,7.7337e-05 526 | CTCF_HUMAN.H10MO.A,34666,0.000782955731485,0.000188768 527 | IKZF1_HUMAN.H10MO.C,15152,0.000224531438244,8.2463e-05 528 | HEN1_HUMAN.H10MO.C,20875,0.000274221332084,6.02603e-05 529 | E2F2_HUMAN.H10MO.B,47533,0.00120323840004,0.000110894 530 | EHF_HUMAN.H10MO.C,49785,0.00109666359466,0.000338912 531 | PROP1_HUMAN.H10MO.D,16345,0.00022890551791,7.60257e-05 532 | KLF3_HUMAN.H10MO.D,32348,0.000593164181693,0.000286341 533 | SNAI2_HUMAN.H10MO.C,10669,0.000196551804181,-1.17719e-05 534 | TBP_HUMAN.H10MO.C,19446,0.000291331116477,8.55327e-05 535 | PAX8_HUMAN.H10MO.D,16658,0.000432770476171,-1.40667e-05 536 | GCR_HUMAN.H10MO.S,6063,0.000206640274672,2.40505e-05 537 | ZN350_HUMAN.H10MO.C,26549,0.000818293539773,0.000157267 538 | HINFP_HUMAN.H10MO.C,24435,0.00036955408771,5.06043e-05 539 | HXC12_HUMAN.H10MO.D,19895,0.000730798702779,0.000142604 540 | ZN143_HUMAN.H10MO.A,42342,0.00097762050213,0.0002262 541 | NR4A3_HUMAN.H10MO.D,9851,0.000394819654132,-2.26498e-06 542 | HESX1_HUMAN.H10MO.D,21295,0.000362489600824,3.09944e-06 543 | PBX1_HUMAN.H10MO.B,15899,0.000334087723226,3.92795e-05 544 | ZEP1_HUMAN.H10MO.D,12391,6.39287784343e-05,-4.58956e-06 545 | GCR_HUMAN.H10MO.A,27756,0.000570567220062,0.000261813 546 | NFIL3_HUMAN.H10MO.C,16018,0.000351608887668,1.19209e-05 547 | SMAD1_HUMAN.H10MO.D,19718,0.000636961789444,9.06885e-05 548 | HXB7_HUMAN.H10MO.C,25636,0.000475921575806,7.58171e-05 549 | PAX4_HUMAN.H10MO.D,18201,0.000598774852568,4.02331e-05 550 | SPIB_HUMAN.H10MO.B,32224,0.000741366584587,0.000220388 551 | PPARA_HUMAN.H10MO.S,5212,0.000209574522752,-3.09944e-06 552 | CEBPA_HUMAN.H10MO.A,13265,0.000532443345076,2.05636e-06 553 | HXA9_HUMAN.H10MO.D,12505,0.000498834519672,4.16636e-05 554 | HIC1_HUMAN.H10MO.C,26519,0.000716497341039,3.69549e-06 555 | THAP1_HUMAN.H10MO.D,56108,0.000857197109385,0.000497431 556 | PPARA_HUMAN.H10MO.C,22816,0.000725353354164,5.57303e-05 557 | FOXO1_HUMAN.H10MO.C,50017,0.00109806869122,0.000821263 558 | GCM1_HUMAN.H10MO.D,24878,0.000744805606776,3.93987e-05 559 | CUX2_HUMAN.H10MO.D,13752,0.000184073652471,3.74615e-05 560 | RORA_HUMAN.H10MO.B,11300,0.000477523621042,-6.04987e-06 561 | ZBT7A_HUMAN.H10MO.D,47435,0.00134986135699,0.000257581 562 | NR1I2_HUMAN.H10MO.S,8259,0.000122796650858,-7.36117e-06 563 | USF1_HUMAN.H10MO.A,27015,0.000311669503691,-9.35793e-06 564 | HXC10_HUMAN.H10MO.D,43995,0.000941970876469,0.000566065 565 | SOX18_HUMAN.H10MO.D,20556,0.000505757346439,7.30753e-05 566 | LMX1B_HUMAN.H10MO.D,41996,0.000772516249553,0.000237137 567 | CEBPZ_HUMAN.H10MO.D,21961,0.000328278193768,3.03388e-05 568 | ZN410_HUMAN.H10MO.D,23282,0.000398660600686,3.40939e-05 569 | EVX1_HUMAN.H10MO.D,34318,0.000420485849757,0.000303298 570 | FOXB1_HUMAN.H10MO.D,17704,0.000419500484261,9.45628e-05 571 | HXC8_HUMAN.H10MO.D,15582,0.000331410680205,-1.07288e-05 572 | PAX2_HUMAN.H10MO.S,30312,0.000467332339253,6.52075e-05 573 | SRBP1_HUMAN.H10MO.B,42870,0.000617870459809,0.000293851 574 | MEIS3_HUMAN.H10MO.D,7811,0.000317078409862,-1.4931e-05 575 | PAX2_HUMAN.H10MO.D,14872,6.93001075498e-05,6.25849e-07 576 | FOXP3_HUMAN.H10MO.D,18272,0.000399447680432,0.000138104 577 | MSX2_HUMAN.H10MO.D,14560,0.000262688704724,3.87728e-05 578 | HXA2_HUMAN.H10MO.D,13062,0.000235201579997,-6.25849e-06 579 | HTF4_HUMAN.H10MO.B,16399,0.000319627687368,-1.10269e-06 580 | ARI3A_HUMAN.H10MO.D,64503,0.00152744508027,0.000655532 581 | OLIG2_HUMAN.H10MO.D,18770,0.000320488873913,1.51694e-05 582 | CR3L1_HUMAN.H10MO.D,23636,0.000691054609874,1.35005e-05 583 | SP2_HUMAN.H10MO.C,84152,0.000857110512915,0.000716597 584 | PIT1_HUMAN.H10MO.C,18782,0.000385255096229,6.83963e-05 585 | TF7L1_HUMAN.H10MO.D,54597,0.00165128300204,0.000294685 586 | FOSL2_HUMAN.H10MO.A,7884,0.000312408374821,-4.91738e-06 587 | ZFHX3_HUMAN.H10MO.D,32692,0.00067547638593,0.000182718 588 | PO3F2_HUMAN.H10MO.D,26282,0.000476298273539,0.000103205 589 | PRD14_HUMAN.H10MO.C,7757,0.000235302703834,1.55866e-05 590 | KLF15_HUMAN.H10MO.D,71788,0.000878186873084,0.000558823 591 | SPIC_HUMAN.H10MO.D,68438,0.00116741313382,0.000753433 592 | ZFX_HUMAN.H10MO.C,52601,0.000901756890581,0.000348538 593 | ESR1_HUMAN.H10MO.A,16376,0.000443273901402,-9.23872e-06 594 | LHX3_HUMAN.H10MO.C,31259,0.000493132960441,9.37581e-05 595 | MTF1_HUMAN.H10MO.C,47544,0.00110166240944,8.57115e-05 596 | LHX2_HUMAN.H10MO.D,14354,0.000367704374984,7.689e-06 597 | FOXH1_HUMAN.H10MO.A,17490,0.000423376605919,5.65052e-05 598 | ZEP2_HUMAN.H10MO.D,31334,0.000730127267078,7.33435e-05 599 | SOX17_HUMAN.H10MO.D,18657,0.000424764659553,6.21676e-05 600 | FIGLA_HUMAN.H10MO.D,13259,0.000195503719635,1.58548e-05 601 | AP2A_HUMAN.H10MO.C,18993,0.000453711107591,4.81308e-05 602 | FOXM1_HUMAN.H10MO.D,30330,0.000689226770216,0.0003905 603 | PO2F2_HUMAN.H10MO.D,19278,0.000539052175541,5.17666e-05 604 | ETV1_HUMAN.H10MO.B,73621,0.00142683315867,0.000467718 605 | GSC2_HUMAN.H10MO.D,50038,0.000895347641015,0.000638932 606 | HXB2_HUMAN.H10MO.D,3536,0.00011938061929,-1.62125e-05 607 | RUNX2_HUMAN.H10MO.B,21508,0.000623971409472,0.000121593 608 | FOXO4_HUMAN.H10MO.C,35784,0.000839757470682,0.000483423 609 | E2F8_HUMAN.H10MO.D,23833,0.000602669166682,0.000114381 610 | BHA15_HUMAN.H10MO.D,24945,0.000367257496826,-8.85129e-06 611 | GCM2_HUMAN.H10MO.D,33974,0.000870634294806,8.21352e-05 612 | ESR2_HUMAN.H10MO.A,19304,0.000695480984707,6.43432e-05 613 | NR2F6_HUMAN.H10MO.D,35301,0.000882113330157,0.000221223 614 | HSF2_HUMAN.H10MO.A,5002,0.000104524758679,-2.5928e-06 615 | LBX2_HUMAN.H10MO.D,10541,0.000126587196926,1.79112e-05 616 | TEAD4_HUMAN.H10MO.A,12001,0.000192646261195,1.28746e-05 617 | PROX1_HUMAN.H10MO.D,66097,0.00100139157652,0.000675231 618 | HEY2_HUMAN.H10MO.D,87336,0.0014925180364,0.000448108 619 | ESR2_HUMAN.H10MO.S,15856,0.000389865933182,-3.18587e-05 620 | AHR_HUMAN.H10MO.B,32582,0.000897363901523,6.48201e-05 621 | KLF1_HUMAN.H10MO.C,80952,0.00145603503166,0.000916749 622 | ZN639_HUMAN.H10MO.D,70151,0.00102481663086,0.000746429 623 | ALX3_HUMAN.H10MO.D,21340,0.00059978101522,7.9602e-05 624 | PLAL1_HUMAN.H10MO.D,45360,0.00107362690224,0.000166684 625 | PRGR_HUMAN.H10MO.C,24141,0.000367555564939,0.000231087 626 | BATF_HUMAN.H10MO.S,6042,3.42845828611e-05,-6.28829e-06 627 | DLX2_HUMAN.H10MO.D,14213,0.000191358078397,1.26362e-05 628 | NF2L2_HUMAN.H10MO.D,9512,0.000410413800155,1.14739e-05 629 | PRGR_HUMAN.H10MO.S,6063,0.000236702460863,2.49445e-05 630 | STAT3_HUMAN.H10MO.A,10904,0.000153289794072,1.00732e-05 631 | HXD12_HUMAN.H10MO.D,16535,0.000462219538255,7.79629e-05 632 | OVOL1_HUMAN.H10MO.C,16429,0.000394538936787,7.24196e-06 633 | BATF_HUMAN.H10MO.A,6930,0.000101059052469,1.53482e-05 634 | MESP1_HUMAN.H10MO.D,19553,0.00045560911722,3.16203e-05 635 | KLF13_HUMAN.H10MO.D,53591,0.00119038210448,0.000470936 636 | ZBTB4_HUMAN.H10MO.D,29533,0.000566967390406,0.000197113 637 | HSFY1_HUMAN.H10MO.D,13812,0.000200533243179,2.69711e-05 638 | TFEB_HUMAN.H10MO.C,17073,0.000359596051122,-1.87755e-06 639 | HXA13_HUMAN.H10MO.C,24540,0.000520942200478,0.000131994 640 | ZBTB4_HUMAN.H10MO.S,48159,0.00150275496228,0.000396073 641 | -------------------------------------------------------------------------------- /feature_importances/SPEID/from_HOCOMOCO_motifs/NHEK_promoters_feature_importance.csv: -------------------------------------------------------------------------------- 1 | Motif Name,Motif Count,AUPR Difference,MS Difference 2 | ETS1_HUMAN.H10MO.C,10731,7.59371227731e-05,7.94828e-05 3 | ATF2_HUMAN.H10MO.B,7415,-0.000612066838566,1.82092e-05 4 | FOXD3_HUMAN.H10MO.D,20116,0.000250584655077,0.000221014 5 | ZBTB6_HUMAN.H10MO.D,12578,0.000878647650941,7.9453e-05 6 | NR1H2_HUMAN.H10MO.D,9384,-0.000121669790673,-2.02656e-05 7 | ETV2_HUMAN.H10MO.D,22655,0.00160048891759,6.08265e-05 8 | ETV3_HUMAN.H10MO.D,11556,0.000985287673364,4.80711e-05 9 | RFX5_HUMAN.H10MO.A,12356,-0.0002262453866,-9.94503e-05 10 | SCRT2_HUMAN.H10MO.D,8009,-0.000412691509187,5.71311e-05 11 | ELF2_HUMAN.H10MO.C,54122,0.00153335715835,8.16286e-05 12 | MLXPL_HUMAN.H10MO.D,16230,-0.000226051700324,0.000211895 13 | PAX6_HUMAN.H10MO.D,9814,0.0012802546193,3.76105e-05 14 | MAX_HUMAN.H10MO.A,13582,0.000232108256921,0.000103056 15 | KLF6_HUMAN.H10MO.D,49087,0.00285126861789,4.6134e-05 16 | HNF1B_HUMAN.H10MO.B,9781,-6.0579376208e-05,9.42945e-05 17 | PBX2_HUMAN.H10MO.C,7044,-0.000203084803259,-2.05636e-06 18 | RXRA_HUMAN.H10MO.C,29682,0.00169132312264,-0.000167102 19 | HESX1_HUMAN.H10MO.D,12022,0.00125826385762,4.81009e-05 20 | NFKB2_HUMAN.H10MO.D,11068,0.000861519257085,8.34763e-05 21 | HSF4_HUMAN.H10MO.D,4168,-9.40622310988e-05,-1.44541e-05 22 | ERR2_HUMAN.H10MO.A,4996,-0.000180349329409,3.8296e-05 23 | ESX1_HUMAN.H10MO.D,11686,-0.000725884433942,0.00018844 24 | AP2B_HUMAN.H10MO.B,49655,0.00124013904007,0.00012961 25 | SPI1_HUMAN.H10MO.A,73458,0.00380972154816,-0.000304252 26 | MYBA_HUMAN.H10MO.D,12335,-0.000102698304906,0.00011766 27 | GFI1B_HUMAN.H10MO.C,5491,0.000429188454564,4.47035e-07 28 | PO6F1_HUMAN.H10MO.D,10432,-0.000164127739463,8.66354e-05 29 | ERR1_HUMAN.H10MO.D,28763,-0.000138761427426,-0.000173628 30 | IRF1_HUMAN.H10MO.A,75189,0.005057519794,1.15931e-05 31 | MAFA_HUMAN.H10MO.D,11089,0.000180133163529,-2.13683e-05 32 | ELF1_HUMAN.H10MO.A,32357,0.000221178466455,0.000201941 33 | SPDEF_HUMAN.H10MO.D,11856,-9.80638415656e-05,0.000128508 34 | GRHL1_HUMAN.H10MO.D,9835,0.000789300623479,5.00679e-05 35 | SCRT1_HUMAN.H10MO.D,6812,0.000285476522037,4.23789e-05 36 | FOXJ2_HUMAN.H10MO.C,21337,0.00130017950998,2.0653e-05 37 | ASCL2_HUMAN.H10MO.D,51032,0.000567472222415,-6.58631e-06 38 | XBP1_HUMAN.H10MO.C,12865,0.000861181485274,-1.42455e-05 39 | FOXF2_HUMAN.H10MO.D,11988,0.000808431743208,2.08318e-05 40 | NFIC_HUMAN.H10MO.A,11083,6.9786772154e-05,5.38528e-05 41 | MYC_HUMAN.H10MO.A,17139,-0.000906086835529,0.000199169 42 | STA5B_HUMAN.H10MO.C,3884,0.000299023574052,2.31266e-05 43 | RELB_HUMAN.H10MO.C,12297,-0.000348817617607,0.000231922 44 | HNF4G_HUMAN.H10MO.C,12919,0.000570780951582,-0.0001432 45 | IRF4_HUMAN.H10MO.C,76058,0.00513745358542,-0.00011453 46 | MEOX2_HUMAN.H10MO.D,12950,-9.7995176325e-05,0.000414222 47 | DBP_HUMAN.H10MO.B,6808,0.000208616343046,5.27203e-05 48 | ANDR_HUMAN.H10MO.A,10583,0.000627493568043,0.000103056 49 | P63_HUMAN.H10MO.A,7807,0.000993215053974,0.000121325 50 | AIRE_HUMAN.H10MO.C,12198,7.01724108351e-05,9.70662e-05 51 | NFYA_HUMAN.H10MO.A,13540,-0.000692162677444,0.000254303 52 | PO5F1_HUMAN.H10MO.A,10613,7.3900451636e-05,-0.000134498 53 | SOX8_HUMAN.H10MO.D,6200,-4.83246103711e-05,3.29316e-05 54 | THA_HUMAN.H10MO.C,10585,0.00056414483929,-1.92523e-05 55 | INSM1_HUMAN.H10MO.C,36329,0.00161385343761,0.00037986 56 | CDX2_HUMAN.H10MO.C,8971,-0.00022602691399,-2.74777e-05 57 | ETS2_HUMAN.H10MO.C,37964,0.00172255532258,5.99325e-05 58 | ZN713_HUMAN.H10MO.D,86181,0.00272895228886,-0.000340194 59 | MYF6_HUMAN.H10MO.C,6507,-2.47234098694e-05,-1.60635e-05 60 | RX_HUMAN.H10MO.D,9857,-0.000446334086277,0.000222772 61 | GATA3_HUMAN.H10MO.C,15840,-0.000922853139383,-0.000187695 62 | ZN148_HUMAN.H10MO.D,81207,0.00139804615386,-0.000186861 63 | PO3F4_HUMAN.H10MO.D,11234,-0.000196195898326,0.000123233 64 | ZN784_HUMAN.H10MO.D,14920,-0.00034742728484,-4.91142e-05 65 | ZN524_HUMAN.H10MO.D,32695,0.00146306301928,0.000238568 66 | STAT6_HUMAN.H10MO.C,5157,0.000487284922862,1.01626e-05 67 | PAX1_HUMAN.H10MO.D,10252,0.000387675231971,0.000140786 68 | PRRX1_HUMAN.H10MO.D,2797,-0.000193419355338,2.38717e-05 69 | ZIC4_HUMAN.H10MO.D,34828,0.00104757147542,0.000263065 70 | CUX1_HUMAN.H10MO.C,9119,6.93661171702e-06,-1.17719e-05 71 | BARH1_HUMAN.H10MO.D,10948,-0.00085645494385,0.000114888 72 | NKX21_HUMAN.H10MO.D,5775,0.00021319019109,2.73585e-05 73 | HXD4_HUMAN.H10MO.D,8277,0.000486688147444,0.000137627 74 | PRRX2_HUMAN.H10MO.C,9186,1.87974037591e-06,0.000104725 75 | PO4F3_HUMAN.H10MO.D,22995,0.00114204346835,0.000478566 76 | HXB6_HUMAN.H10MO.D,9476,0.000291006181801,3.74317e-05 77 | NDF2_HUMAN.H10MO.D,15520,6.05738180341e-05,-1.75536e-05 78 | HXD3_HUMAN.H10MO.D,9232,-0.000452183038424,0.000194371 79 | SRF_HUMAN.H10MO.A,17295,-0.000625085544369,4.15742e-05 80 | ITF2_HUMAN.H10MO.B,7795,0.000124630129036,0.000123739 81 | MYCN_HUMAN.H10MO.B,15909,0.000174403790685,0.000267476 82 | TFE2_HUMAN.H10MO.C,6092,0.00177838260005,6.1363e-05 83 | FLI1_HUMAN.H10MO.A,31103,0.0010918696196,0.000182122 84 | PURA_HUMAN.H10MO.D,59182,0.00116155797102,-0.000172406 85 | ETV7_HUMAN.H10MO.D,10040,0.00141133584704,0.000192702 86 | NFIA_HUMAN.H10MO.C,11779,-0.000128627128904,-3.38256e-05 87 | LHX8_HUMAN.H10MO.D,6117,-0.000322927045001,1.75834e-06 88 | COE1_HUMAN.H10MO.A,12427,0.000188505417116,0.000170678 89 | KLF8_HUMAN.H10MO.C,19822,0.000263325134634,9.84669e-05 90 | P53_HUMAN.H10MO.B,7396,0.000221388821873,9.65893e-05 91 | STA5A_HUMAN.H10MO.B,4065,-0.000406697229685,0.000100076 92 | FOS_HUMAN.H10MO.A,3984,0.000233918910488,6.90818e-05 93 | ZSC16_HUMAN.H10MO.D,5734,2.44703118357e-05,-2.18749e-05 94 | TBX21_HUMAN.H10MO.D,8156,0.000532951102902,-9.41455e-05 95 | GATA5_HUMAN.H10MO.D,9390,-0.000422148105741,8.55327e-06 96 | HES1_HUMAN.H10MO.D,22252,0.00194474246158,0.000128806 97 | TFCP2_HUMAN.H10MO.D,14745,0.000693557135167,0.000108361 98 | TGIF1_HUMAN.H10MO.S,3613,-0.000418816421613,-3.77893e-05 99 | BACH1_HUMAN.H10MO.A,7431,-0.000146090433411,6.24061e-05 100 | GSX1_HUMAN.H10MO.D,11873,-0.0010308711022,0.00029707 101 | TGIF1_HUMAN.H10MO.D,8006,-0.000226053649277,-8.04663e-05 102 | SOX10_HUMAN.H10MO.D,10973,-0.000353904975237,0.000126719 103 | RARB_HUMAN.H10MO.D,9319,-7.93697721495e-05,-3.13222e-05 104 | FOXC2_HUMAN.H10MO.D,6871,0.000634935661764,-3.91304e-05 105 | NFAC4_HUMAN.H10MO.C,17835,0.000579495971721,-7.88867e-05 106 | TBX5_HUMAN.H10MO.D,7971,0.000645214098187,-0.000179231 107 | UNC4_HUMAN.H10MO.D,13272,-0.00120248453589,0.000245005 108 | HXA11_HUMAN.H10MO.D,14974,0.00138154588945,0.000216246 109 | MAFG_HUMAN.H10MO.S,2730,0.000269705100765,-4.58956e-05 110 | DRGX_HUMAN.H10MO.D,11338,0.000329804545339,0.000197113 111 | BHE41_HUMAN.H10MO.D,14486,0.000919409210726,0.000211746 112 | GLIS3_HUMAN.H10MO.D,41915,0.00170213219508,0.000137985 113 | GBX2_HUMAN.H10MO.D,11770,-0.00103736016671,0.000281751 114 | ELF3_HUMAN.H10MO.D,13265,0.00013910830605,0.000158042 115 | TF2LX_HUMAN.H10MO.D,5191,-1.98647969997e-05,-5.73993e-05 116 | FOXD1_HUMAN.H10MO.D,8744,-0.000724719227241,6.62208e-05 117 | CREB1_HUMAN.H10MO.A,10565,-0.000647063419914,-3.24547e-05 118 | EVX2_HUMAN.H10MO.A,16240,-0.000873622837157,0.000126392 119 | BSH_HUMAN.H10MO.D,11854,0.000140938684563,0.00029549 120 | ZIC3_HUMAN.H10MO.C,25071,-0.000532209470667,0.000238568 121 | ISL2_HUMAN.H10MO.D,5668,0.000237814671017,6.59227e-05 122 | ZIC2_HUMAN.H10MO.C,49146,-0.000621445464945,0.000394702 123 | EGR2_HUMAN.H10MO.C,83090,0.00176234270236,-5.48959e-05 124 | MAFK_HUMAN.H10MO.S,3805,-0.000542059118066,7.27177e-05 125 | FOXQ1_HUMAN.H10MO.C,17583,0.000445652202473,8.34465e-05 126 | ERR3_HUMAN.H10MO.B,3500,0.00029143803168,-3.06368e-05 127 | GMEB2_HUMAN.H10MO.D,8547,0.0006828065115,8.16286e-05 128 | JDP2_HUMAN.H10MO.D,8738,0.000499647602213,-3.97563e-05 129 | SMAD3_HUMAN.H10MO.C,8885,0.000471946973372,2.61068e-05 130 | ETV5_HUMAN.H10MO.D,11090,0.000261451150767,0.00010249 131 | NKX28_HUMAN.H10MO.C,3899,-0.000427593371843,7.33733e-05 132 | OLIG3_HUMAN.H10MO.D,8879,3.79766043795e-05,-4.1008e-05 133 | LMX1A_HUMAN.H10MO.D,23789,-0.000583938618213,0.000261217 134 | HNF1A_HUMAN.H10MO.A,10071,-0.000443619425694,0.000147134 135 | NANOG_HUMAN.H10MO.S,9996,4.34687053034e-05,8.71718e-05 136 | CREM_HUMAN.H10MO.C,9867,0.00025402143335,-5.28991e-05 137 | MAFG_HUMAN.H10MO.C,4651,0.000818326922173,-4.54187e-05 138 | PRDM1_HUMAN.H10MO.C,35605,0.00312571655144,-0.000177205 139 | NANOG_HUMAN.H10MO.A,8970,-0.000365821033159,0.000169665 140 | ARI5B_HUMAN.H10MO.C,8338,0.000343056642783,9.06885e-05 141 | MEIS2_HUMAN.H10MO.B,5469,0.000196465822219,-4.79221e-05 142 | ZN384_HUMAN.H10MO.C,19287,-0.000879836692807,-5.3376e-05 143 | PITX2_HUMAN.H10MO.D,11416,0.000321369234817,9.31621e-05 144 | ATF1_HUMAN.H10MO.B,8042,7.96679608819e-05,2.84314e-05 145 | STAT2_HUMAN.H10MO.B,51260,0.0054514648206,3.66271e-05 146 | ELK3_HUMAN.H10MO.D,13945,0.000602974359891,0.00010556 147 | NOTO_HUMAN.H10MO.D,7688,-0.000777775482545,0.000210047 148 | SP4_HUMAN.H10MO.D,82066,0.000481888726589,-0.000343949 149 | ZBT18_HUMAN.H10MO.D,6332,-0.000719363868259,5.48363e-05 150 | ELF5_HUMAN.H10MO.D,13074,0.00235341825626,0.000251919 151 | NKX61_HUMAN.H10MO.D,16781,-0.000132583561368,0.000246435 152 | PKNX1_HUMAN.H10MO.D,6166,-0.000730567367749,-1.49906e-05 153 | MSX1_HUMAN.H10MO.D,11830,0.000123806393045,0.000160575 154 | DLX4_HUMAN.H10MO.D,14019,-8.25937700071e-05,0.000230938 155 | REST_HUMAN.H10MO.A,13363,-0.00180408830537,0.000174552 156 | RARA_HUMAN.H10MO.S,2623,-0.00105530660483,8.58307e-06 157 | FOSL1_HUMAN.H10MO.A,3587,0.000421611088851,4.94719e-05 158 | FUBP1_HUMAN.H10MO.D,40463,0.000544851140553,0.000180364 159 | REL_HUMAN.H10MO.C,13876,0.00018603841845,0.000188112 160 | PO4F1_HUMAN.H10MO.D,21432,-0.000121823780571,0.000386119 161 | PHX2A_HUMAN.H10MO.D,10939,-0.000644364165237,0.000209063 162 | PO2F1_HUMAN.H10MO.B,9721,6.44818914675e-06,3.79384e-05 163 | HXB8_HUMAN.H10MO.C,12260,0.000520990068889,0.000115782 164 | RFX2_HUMAN.H10MO.C,18239,-0.000480544756264,7.07507e-05 165 | VAX1_HUMAN.H10MO.D,12743,0.00142708762469,0.000291675 166 | EOMES_HUMAN.H10MO.D,12063,0.000655976930866,2.94149e-05 167 | ATF7_HUMAN.H10MO.D,8363,-0.000107626140607,-1.40369e-05 168 | ISL1_HUMAN.H10MO.D,7462,0.000771418138914,0.000169754 169 | LEF1_HUMAN.H10MO.C,4915,2.88620666972e-05,-4.01437e-05 170 | SRBP2_HUMAN.H10MO.B,71520,0.00235699664194,-0.00011456 171 | ATF6A_HUMAN.H10MO.B,21269,-1.28960396562e-06,-0.000248373 172 | TFDP1_HUMAN.H10MO.S,32594,-0.00167528164712,0.00014925 173 | ZN423_HUMAN.H10MO.D,13926,0.000875309998414,0.000178516 174 | NR4A1_HUMAN.H10MO.C,8449,0.000715002925052,4.52995e-06 175 | NFAC1_HUMAN.H10MO.A,12520,-0.000372046023634,0.000169724 176 | PPARG_HUMAN.H10MO.A,39923,0.00100984907019,-0.000370204 177 | TFDP1_HUMAN.H10MO.D,55270,0.00181959460271,-0.000164002 178 | RFX4_HUMAN.H10MO.D,6364,-0.000737849840282,-2.22325e-05 179 | CEBPG_HUMAN.H10MO.C,10044,0.000374900725309,5.39422e-05 180 | NFAC1_HUMAN.H10MO.S,8928,0.00028821605053,0.000294805 181 | PPARG_HUMAN.H10MO.S,7852,-4.45900660596e-05,-3.59714e-05 182 | SOX2_HUMAN.H10MO.B,11636,-9.76738243926e-05,-8.81255e-05 183 | TBX1_HUMAN.H10MO.D,87223,0.00208596334371,-8.68142e-05 184 | SOX7_HUMAN.H10MO.D,6675,0.000179072898398,6.15716e-05 185 | GATA1_HUMAN.H10MO.S,12967,0.000446613979863,-3.16799e-05 186 | MYOD1_HUMAN.H10MO.C,7764,0.000241042679733,1.77324e-05 187 | WT1_HUMAN.H10MO.D,38318,-0.00107645471302,-8.90195e-05 188 | BHE22_HUMAN.H10MO.D,10125,3.76207195859e-05,6.10054e-05 189 | GATA1_HUMAN.H10MO.A,16458,0.000513024623369,-0.00020504 190 | THAP1_HUMAN.H10MO.D,80922,0.000539584633625,-3.87132e-05 191 | E2F7_HUMAN.H10MO.D,27173,-0.000413827060389,5.1558e-05 192 | HAND1_HUMAN.H10MO.D,5036,-0.000100109431917,-1.96695e-05 193 | P5F1B_HUMAN.H10MO.D,10721,0.000633268346202,0.000108629 194 | HXD8_HUMAN.H10MO.D,7760,0.000107793038727,2.21133e-05 195 | ELK1_HUMAN.H10MO.A,12193,-0.00183204767916,5.85616e-05 196 | FOSB_HUMAN.H10MO.C,3824,0.000575044761063,3.67761e-05 197 | TBX2_HUMAN.H10MO.D,6737,-4.11936642418e-05,2.38419e-07 198 | FOXO3_HUMAN.H10MO.B,18651,-0.000886641296792,0.000148982 199 | RUNX3_HUMAN.H10MO.C,8518,0.000171790897607,7.82907e-05 200 | NKX23_HUMAN.H10MO.D,6731,2.62553525356e-05,3.29912e-05 201 | NKX62_HUMAN.H10MO.D,11212,0.000356108226231,0.000213712 202 | ESR1_HUMAN.H10MO.S,10834,-0.000163661764435,4.59254e-05 203 | MAZ_HUMAN.H10MO.A,58083,0.000979355289817,-0.000198781 204 | HXA7_HUMAN.H10MO.D,4428,-0.000312415766165,2.08616e-05 205 | CRX_HUMAN.H10MO.C,9623,0.000279403580581,-0.000101894 206 | SHOX_HUMAN.H10MO.D,14082,-7.86944904929e-05,0.000269383 207 | NOBOX_HUMAN.H10MO.C,7068,-0.00074587689529,0.000135601 208 | MZF1_HUMAN.H10MO.D,82810,0.00311373274719,0.000574678 209 | STAT1_HUMAN.H10MO.A,11680,0.000119438783388,0.000209421 210 | CTCFL_HUMAN.H10MO.A,24906,-5.08574599476e-05,0.000258625 211 | CEBPD_HUMAN.H10MO.B,10864,-3.28820125119e-05,0.000118494 212 | TF7L2_HUMAN.H10MO.A,13378,-0.000704377852168,8.12113e-05 213 | ID4_HUMAN.H10MO.D,8294,0.000337494307439,3.52561e-05 214 | PAX3_HUMAN.H10MO.D,8718,-0.000400615934567,3.71337e-05 215 | STAT1_HUMAN.H10MO.S,7089,0.000193620143436,7.23898e-05 216 | PAX7_HUMAN.H10MO.D,9556,-0.000182423645794,9.85861e-05 217 | RFX1_HUMAN.H10MO.C,8298,0.000699187026123,8.32975e-05 218 | RXRG_HUMAN.H10MO.B,10267,-0.000326059998309,-7.89762e-06 219 | TBX20_HUMAN.H10MO.D,23593,0.0017204528808,-0.000111133 220 | MUSC_HUMAN.H10MO.D,19321,0.00100639533685,4.71473e-05 221 | ZSCA4_HUMAN.H10MO.D,10147,0.000322932278335,-1.65701e-05 222 | VSX2_HUMAN.H10MO.D,5189,0.000305192971958,7.96616e-05 223 | NFAT5_HUMAN.H10MO.D,8499,2.34991151732e-05,-5.36442e-07 224 | BCL6_HUMAN.H10MO.C,3535,0.000346133476485,1.85072e-05 225 | ZIC1_HUMAN.H10MO.B,61665,0.00186828833343,0.000367343 226 | MEF2D_HUMAN.H10MO.C,8762,0.000277204084285,-3.35574e-05 227 | SP3_HUMAN.H10MO.B,66566,0.000290578827423,-0.000116765 228 | NR1H4_HUMAN.H10MO.C,9366,-0.000933361074668,5.14388e-05 229 | E2F4_HUMAN.H10MO.A,40669,0.00117799148336,6.02603e-05 230 | E2F5_HUMAN.H10MO.B,22566,0.000317038987557,0.00024429 231 | HLF_HUMAN.H10MO.C,9701,0.00134092757941,3.80874e-05 232 | TBX4_HUMAN.H10MO.D,7139,7.73580393746e-05,-0.00015381 233 | HXD13_HUMAN.H10MO.D,7502,-0.00023860563739,7.21216e-06 234 | HME1_HUMAN.H10MO.D,11089,-0.000865072329508,0.000259012 235 | TYY1_HUMAN.H10MO.A,15981,0.000746556262483,-1.37091e-05 236 | FOXF1_HUMAN.H10MO.D,24665,-0.000874661355514,0.000103951 237 | EMX1_HUMAN.H10MO.D,9283,-4.46939825307e-05,0.000289202 238 | ZN232_HUMAN.H10MO.D,8183,0.00085672919874,-7.7486e-06 239 | BC11A_HUMAN.H10MO.C,67510,0.00275615164569,-0.000184923 240 | SOX13_HUMAN.H10MO.D,14942,-0.000826611876045,0.000134349 241 | NKX32_HUMAN.H10MO.C,10382,0.00155702039524,-7.34329e-05 242 | CREB3_HUMAN.H10MO.D,15583,-0.000320529271606,-0.000115722 243 | PO2F3_HUMAN.H10MO.D,7624,-0.000655519900093,9.53078e-05 244 | HMX2_HUMAN.H10MO.D,9138,-0.000862931272,9.99272e-05 245 | SNAI1_HUMAN.H10MO.C,8348,0.000274544502362,0.000141293 246 | HXC6_HUMAN.H10MO.D,18250,0.00105834188473,0.000198692 247 | P63_HUMAN.H10MO.S,7034,0.000227489915116,7.56681e-05 248 | VAX2_HUMAN.H10MO.D,7918,0.000623362358913,4.60744e-05 249 | TBX19_HUMAN.H10MO.D,6175,0.000797361097215,1.71959e-05 250 | TLX1_HUMAN.H10MO.S,10928,0.000593328747363,0.000183403 251 | BATF3_HUMAN.H10MO.D,13339,5.77412725731e-05,0.000153124 252 | HXB1_HUMAN.H10MO.D,13316,0.00049286172783,0.000209957 253 | PDX1_HUMAN.H10MO.C,5013,1.98225394169e-05,0.000129402 254 | TF65_HUMAN.H10MO.A,12547,0.000412440171379,0.000137955 255 | LHX6_HUMAN.H10MO.D,9058,-0.000125711428558,8.21948e-05 256 | E4F1_HUMAN.H10MO.D,11949,-0.000416607049053,-0.000137627 257 | BHE23_HUMAN.H10MO.D,15913,0.000529864199734,-1.78218e-05 258 | SOX3_HUMAN.H10MO.D,7655,-0.00119255585386,0.000111789 259 | PEBB_HUMAN.H10MO.C,9821,0.000882502550907,3.84748e-05 260 | TLX1_HUMAN.H10MO.D,8010,0.000909274768391,3.70145e-05 261 | NFYB_HUMAN.H10MO.A,12399,-0.00106827934138,0.000168383 262 | ESR1_HUMAN.H10MO.A,7363,-0.000559719124214,8.85129e-06 263 | EGR4_HUMAN.H10MO.D,85424,0.000881864319827,6.3926e-05 264 | HME2_HUMAN.H10MO.D,10022,0.000904990946274,1.00732e-05 265 | BARX1_HUMAN.H10MO.D,13820,0.0014448130789,0.000375271 266 | HXC11_HUMAN.H10MO.D,9499,-0.0008968141903,-4.27961e-05 267 | HXB3_HUMAN.H10MO.D,11201,-0.00111095017742,0.00036484 268 | OLIG1_HUMAN.H10MO.D,9901,0.000551730264938,9.71556e-06 269 | CR3L2_HUMAN.H10MO.D,18341,-2.24078022792e-06,-5.74589e-05 270 | ALX3_HUMAN.H10MO.D,11181,0.000724369271681,0.000135124 271 | CDC5L_HUMAN.H10MO.D,12901,-3.95691052599e-05,3.96073e-05 272 | MEOX1_HUMAN.H10MO.D,10548,-0.000231725244837,0.000171423 273 | EPAS1_HUMAN.H10MO.D,56060,0.00179505219287,0.000179917 274 | CEBPB_HUMAN.H10MO.A,8093,0.000395478283762,9.79304e-05 275 | IRF3_HUMAN.H10MO.C,89353,0.00644043133687,2.53916e-05 276 | KAISO_HUMAN.H10MO.A,33169,0.00110910674146,-3.38554e-05 277 | RUNX1_HUMAN.H10MO.A,8374,0.000446785504968,4.8399e-05 278 | PO3F3_HUMAN.H10MO.D,24060,0.000372330707673,0.000180185 279 | IRX2_HUMAN.H10MO.D,43239,0.000807312282219,8.46982e-05 280 | KAISO_HUMAN.H10MO.S,9013,0.000217195532023,8.06153e-05 281 | MYB_HUMAN.H10MO.C,9849,0.0004876182605,-8.63373e-05 282 | GFI1_HUMAN.H10MO.C,4997,0.000172424348918,-1.68383e-05 283 | ETV4_HUMAN.H10MO.B,10770,0.000465869169611,-1.60038e-05 284 | NR5A2_HUMAN.H10MO.C,5692,0.000437456246842,0.000132829 285 | HBP1_HUMAN.H10MO.D,6419,-0.000238080757411,5.75781e-05 286 | ARX_HUMAN.H10MO.D,10872,0.000237741989343,0.000149667 287 | EMX2_HUMAN.H10MO.D,10406,0.000838840797406,0.000225782 288 | ARNT2_HUMAN.H10MO.D,47796,0.0029142345622,-0.000146836 289 | ZN333_HUMAN.H10MO.D,12950,2.70728072449e-06,0.000122011 290 | NR2E1_HUMAN.H10MO.D,7208,0.000462212203066,3.1352e-05 291 | FOXA3_HUMAN.H10MO.C,19624,0.000395707552147,3.17991e-05 292 | BARX2_HUMAN.H10MO.D,11768,0.000935295952467,0.000187218 293 | TEF_HUMAN.H10MO.D,9708,-0.00011698773688,1.60635e-05 294 | ONEC3_HUMAN.H10MO.D,9882,0.000818824128062,8.74698e-05 295 | ONEC2_HUMAN.H10MO.D,17675,0.00100001327666,0.000181019 296 | ZN282_HUMAN.H10MO.D,31807,0.00118198925019,-0.000190526 297 | HXB13_HUMAN.H10MO.D,9689,-2.27632798819e-05,9.20892e-06 298 | MIXL1_HUMAN.H10MO.D,8218,-5.66872017413e-05,0.000199586 299 | FOXP2_HUMAN.H10MO.A,9151,-0.000546648504962,2.74479e-05 300 | TEAD3_HUMAN.H10MO.D,6705,-0.000510505577392,1.508e-05 301 | TWST1_HUMAN.H10MO.D,10698,0.00105032257417,0.000129491 302 | OTX1_HUMAN.H10MO.D,6434,8.1596805463e-05,-2.92957e-05 303 | NR6A1_HUMAN.H10MO.B,5709,-8.86605068157e-05,3.43621e-05 304 | HXD9_HUMAN.H10MO.D,11558,0.000637786920959,0.000108629 305 | BRAC_HUMAN.H10MO.D,4893,-0.000146479399401,3.94285e-05 306 | ERF_HUMAN.H10MO.D,8034,4.35843044064e-05,8.36551e-05 307 | FOXA1_HUMAN.H10MO.A,9428,0.000263802299882,-8.76188e-06 308 | FOXJ3_HUMAN.H10MO.A,37368,0.00202294484215,0.000272572 309 | LHX3_HUMAN.H10MO.C,17434,-0.000698050755737,0.00031808 310 | MYOG_HUMAN.H10MO.D,5756,0.00011085924071,1.21593e-05 311 | KLF13_HUMAN.H10MO.D,71868,0.00142554933133,0.000180155 312 | MAFB_HUMAN.H10MO.D,5420,0.000131384490826,-4.98891e-05 313 | GLI1_HUMAN.H10MO.C,18172,0.000391698664714,0.000187635 314 | NGN2_HUMAN.H10MO.D,13162,0.00113476198409,-3.60012e-05 315 | PLAG1_HUMAN.H10MO.D,90614,0.00405467252287,0.000195444 316 | E2F1_HUMAN.H10MO.A,66201,0.00257882904226,0.000341058 317 | YBOX1_HUMAN.H10MO.D,15641,-0.00063728478307,0.000164092 318 | PLAG1_HUMAN.H10MO.S,53611,0.000515771117864,0.000274599 319 | HMX1_HUMAN.H10MO.D,10781,-9.03078422536e-05,0.000215024 320 | SOX1_HUMAN.H10MO.D,6429,-1.83491569836e-05,1.40071e-06 321 | TBX15_HUMAN.H10MO.D,69671,0.000568928620154,-1.71065e-05 322 | PPARD_HUMAN.H10MO.D,17329,-0.000462041667994,-7.85887e-05 323 | RFX3_HUMAN.H10MO.B,5677,-0.000268392492228,1.78814e-05 324 | USF2_HUMAN.H10MO.A,18764,-0.000420524078708,0.00013411 325 | RREB1_HUMAN.H10MO.D,83842,0.000942112169793,9.17315e-05 326 | MAFF_HUMAN.H10MO.A,3776,0.000205670848493,2.33054e-05 327 | FOXD2_HUMAN.H10MO.D,15112,0.00104339395845,0.000212491 328 | E2F3_HUMAN.H10MO.B,29707,0.00124006445,0.000102043 329 | PO4F2_HUMAN.H10MO.D,10626,0.000107498023415,0.000258058 330 | ALX4_HUMAN.H10MO.D,10992,-0.000272626218911,0.000105083 331 | THA_HUMAN.H10MO.S,5615,-0.000288817567085,-9.11951e-06 332 | MAFK_HUMAN.H10MO.A,4233,0.000444240903263,-2.76864e-05 333 | HIC2_HUMAN.H10MO.D,12488,-0.000646126435368,2.86102e-05 334 | NFIA_HUMAN.H10MO.S,2676,-0.000282277957652,1.29044e-05 335 | STAT4_HUMAN.H10MO.D,5940,0.000181961558557,4.8548e-05 336 | TGIF2_HUMAN.H10MO.D,4953,-0.000238821937977,7.13468e-05 337 | MCR_HUMAN.H10MO.D,10041,0.000479405923481,-0.000151664 338 | PITX3_HUMAN.H10MO.D,14200,-0.000183618926333,0.000130534 339 | SHOX2_HUMAN.H10MO.D,13871,0.000309333541978,0.000261664 340 | NR1D1_HUMAN.H10MO.C,6451,-0.00193114605746,-2.14279e-05 341 | MEIS1_HUMAN.H10MO.C,4740,0.000642708571151,-5.95748e-05 342 | ZN740_HUMAN.H10MO.D,93591,0.00104843991579,0.000236213 343 | PRDM4_HUMAN.H10MO.D,12426,0.000875706317365,7.86185e-05 344 | GSC_HUMAN.H10MO.D,14071,0.000549671922528,6.74129e-05 345 | DMBX1_HUMAN.H10MO.D,14010,-0.000732449181109,0.000192314 346 | PO6F2_HUMAN.H10MO.D,6592,0.000153844849037,0.000139534 347 | SOX5_HUMAN.H10MO.C,15386,0.000446804806319,5.00381e-05 348 | NRL_HUMAN.H10MO.D,5229,6.33377053706e-05,3.62396e-05 349 | ATF3_HUMAN.H10MO.A,7644,-0.000408849183454,3.25143e-05 350 | SP1_HUMAN.H10MO.C,55269,0.00119023288304,-0.000176013 351 | ETV6_HUMAN.H10MO.D,16520,0.000513503894263,0.000128329 352 | IRF8_HUMAN.H10MO.D,49065,0.00255525595284,-2.79844e-05 353 | SP1_HUMAN.H10MO.S,60642,0.00210776742064,-9.16421e-05 354 | SPZ1_HUMAN.H10MO.D,25977,-0.000126844720544,0.000434607 355 | BPTF_HUMAN.H10MO.D,32672,0.00169924801583,0.0001131 356 | PHX2B_HUMAN.H10MO.D,10980,-5.84363091813e-05,0.000199616 357 | HSF1_HUMAN.H10MO.A,9307,0.000175083589474,-2.42293e-05 358 | CXXC1_HUMAN.H10MO.D,13018,-0.00150272596287,4.72367e-05 359 | MYBB_HUMAN.H10MO.D,6502,-0.00103009557383,6.44028e-05 360 | GATA6_HUMAN.H10MO.B,16516,0.000534478626073,-0.000147551 361 | FOXK1_HUMAN.H10MO.D,29829,0.00114875866644,0.00021857 362 | BMAL1_HUMAN.H10MO.C,13825,0.000258287977805,0.000159025 363 | GLI2_HUMAN.H10MO.B,10954,-3.90396336546e-05,7.68006e-05 364 | DLX6_HUMAN.H10MO.D,15362,-0.000302049777951,0.000331551 365 | JUND_HUMAN.H10MO.A,4024,0.000297635011887,8.57711e-05 366 | HXD11_HUMAN.H10MO.D,15442,-0.000926199553775,0.000108719 367 | TFE3_HUMAN.H10MO.C,9130,-0.000649847045237,1.39773e-05 368 | TYY2_HUMAN.H10MO.D,26074,0.00141800839929,0.000239223 369 | HIF1A_HUMAN.H10MO.A,23830,0.00147838929626,0.000143856 370 | NKX25_HUMAN.H10MO.C,2074,-7.34569138814e-05,1.16527e-05 371 | KLF14_HUMAN.H10MO.D,55514,0.00119507186834,6.94096e-05 372 | NDF1_HUMAN.H10MO.C,13024,-0.000944571997004,0.000259638 373 | NR0B1_HUMAN.H10MO.D,65281,0.00129360204434,-0.000148386 374 | LHX9_HUMAN.H10MO.D,13138,-0.000414059688766,0.000211716 375 | HXA10_HUMAN.H10MO.C,10808,-5.32788229999e-05,0.000111341 376 | NRF1_HUMAN.H10MO.A,49607,0.00346862208348,0.000262082 377 | NF2L1_HUMAN.H10MO.C,2730,0.000559393937406,-2.47061e-05 378 | ATOH1_HUMAN.H10MO.D,9514,-0.000365941742437,1.00136e-05 379 | ZBED1_HUMAN.H10MO.D,6769,-6.10240530065e-05,-7.28667e-05 380 | GATA4_HUMAN.H10MO.B,10585,0.000402931560157,-9.49502e-05 381 | SOX9_HUMAN.H10MO.B,18524,0.000837044092361,3.39746e-06 382 | LHX4_HUMAN.H10MO.D,10430,-0.000812924894865,0.000285476 383 | NR4A2_HUMAN.H10MO.C,6430,-0.000109659241041,-7.59959e-06 384 | MNX1_HUMAN.H10MO.D,32518,0.00159208081855,-1.22786e-05 385 | MGAP_HUMAN.H10MO.D,6724,0.000352893992693,-0.00017643 386 | AP2D_HUMAN.H10MO.D,69534,0.00327970817625,-5.43296e-05 387 | BCL6B_HUMAN.H10MO.D,5143,0.000830908976127,2.5332e-05 388 | PTF1A_HUMAN.H10MO.C,12984,0.000933987623192,0.0001176 389 | SMRC1_HUMAN.H10MO.D,4948,-0.000116368061388,1.94609e-05 390 | HMGA1_HUMAN.H10MO.D,11425,0.00067698442388,1.54674e-05 391 | JUNB_HUMAN.H10MO.C,5408,0.000229329491352,6.43134e-05 392 | CENPB_HUMAN.H10MO.D,16904,0.000188464301897,0.000139862 393 | FOXA2_HUMAN.H10MO.A,9604,0.000153290303117,1.06692e-05 394 | IRF5_HUMAN.H10MO.D,70312,0.00508836251245,-0.000279367 395 | NFYC_HUMAN.H10MO.B,15016,0.000275642452877,0.000198931 396 | ARNT_HUMAN.H10MO.B,17672,0.000702157153768,0.000345141 397 | RORG_HUMAN.H10MO.C,8118,-0.000275632051534,6.83367e-05 398 | RAX2_HUMAN.H10MO.D,11304,-0.000976601778442,0.000183731 399 | COT1_HUMAN.H10MO.B,5781,0.000309515385043,-2.30372e-05 400 | THB_HUMAN.H10MO.S,6147,-0.000778945398181,3.61204e-05 401 | ZEB1_HUMAN.H10MO.B,8794,-0.000515429331109,-3.57628e-06 402 | SMAD4_HUMAN.H10MO.C,10907,0.000431427945672,-0.000121772 403 | THB_HUMAN.H10MO.C,15586,-0.000837599423985,0.000272065 404 | KLF12_HUMAN.H10MO.D,29162,0.00106544071776,-7.77841e-06 405 | SOX21_HUMAN.H10MO.D,7066,-0.000620693593582,5.46873e-05 406 | E2F6_HUMAN.H10MO.C,57685,-0.000609948662213,9.5576e-05 407 | MBD2_HUMAN.H10MO.B,55894,0.00150749280433,0.00019151 408 | HXD10_HUMAN.H10MO.D,15520,0.000625131417392,0.000171423 409 | P73_HUMAN.H10MO.S,11508,1.87332899141e-05,2.52128e-05 410 | SMAD2_HUMAN.H10MO.C,10202,-0.000184503184867,-0.000148535 411 | EGR1_HUMAN.H10MO.A,69055,-0.00185249257858,-0.000169039 412 | JUN_HUMAN.H10MO.A,3830,-0.000341242077878,2.69413e-05 413 | CEBPE_HUMAN.H10MO.A,10036,8.80706397607e-05,0.000160366 414 | RARG_HUMAN.H10MO.C,20705,-0.000282306928741,5.59986e-05 415 | IRX3_HUMAN.H10MO.D,11432,0.000426530094084,-2.64049e-05 416 | EGR3_HUMAN.H10MO.D,39949,0.000777638859778,0.000147492 417 | P73_HUMAN.H10MO.A,9102,0.000576414948307,5.67436e-05 418 | HNF6_HUMAN.H10MO.C,8356,0.000576050290375,3.08156e-05 419 | DUXA_HUMAN.H10MO.D,6172,-3.61266978999e-07,6.20782e-05 420 | FOXL1_HUMAN.H10MO.D,57807,0.00181575452821,0.000323713 421 | EGR1_HUMAN.H10MO.S,99441,0.0040932040707,-2.13087e-05 422 | HLTF_HUMAN.H10MO.D,9930,-0.000509337836089,5.1111e-05 423 | RARG_HUMAN.H10MO.S,2623,-0.00114889080047,-9.35793e-06 424 | COT2_HUMAN.H10MO.A,14221,-0.000281092035758,-8.29995e-05 425 | UBIP1_HUMAN.H10MO.D,6051,0.000386336893021,-0.000148833 426 | MAF_HUMAN.H10MO.B,3335,-0.000125397624276,2.66433e-05 427 | HXC13_HUMAN.H10MO.D,9905,0.000708637653325,4.72367e-05 428 | DDIT3_HUMAN.H10MO.C,10022,0.000180488782131,6.43134e-05 429 | CPEB1_HUMAN.H10MO.D,51049,0.00113598039818,-0.000220269 430 | COT2_HUMAN.H10MO.S,19058,-0.000166341009892,-0.000153601 431 | PITX1_HUMAN.H10MO.D,16312,0.000293716130884,-5.81145e-05 432 | ENOA_HUMAN.H10MO.A,16659,0.00071496141391,6.24955e-05 433 | ZKSC3_HUMAN.H10MO.D,4401,0.000456996344665,-4.1604e-05 434 | RXRB_HUMAN.H10MO.C,5577,1.49285102454e-05,-5.48661e-05 435 | NFE2_HUMAN.H10MO.B,4560,-8.76533919203e-05,4.1157e-05 436 | TAL1_HUMAN.H10MO.A,10650,0.000298736002061,-2.49147e-05 437 | SOX11_HUMAN.H10MO.D,6298,-0.000420146308568,7.19726e-05 438 | GSX2_HUMAN.H10MO.D,14157,-0.000308195485002,0.000256419 439 | TAL1_HUMAN.H10MO.S,8227,-9.32051464611e-05,0.000124574 440 | BRCA1_HUMAN.H10MO.D,12398,0.000923118761241,3.56734e-05 441 | FEV_HUMAN.H10MO.C,12939,0.00116377632122,5.57303e-06 442 | TFAP4_HUMAN.H10MO.C,10329,0.00010329378718,0.00012356 443 | VSX1_HUMAN.H10MO.D,11795,0.000483379558783,0.000275791 444 | PBX3_HUMAN.H10MO.B,7272,0.000451598763017,0.000151396 445 | MLX_HUMAN.H10MO.D,11003,0.000481615189699,0.000172824 446 | IRF7_HUMAN.H10MO.C,31210,0.00386374902887,0.000184029 447 | PO3F1_HUMAN.H10MO.C,10245,-0.000860432264101,5.95152e-05 448 | SRY_HUMAN.H10MO.B,30990,0.0021235237585,0.000190318 449 | HMX3_HUMAN.H10MO.D,8339,0.0012969090537,7.23004e-05 450 | SOX4_HUMAN.H10MO.C,27349,0.00178266308243,-0.000177056 451 | CDX1_HUMAN.H10MO.C,10397,0.000437084615075,9.79006e-05 452 | RARA_HUMAN.H10MO.C,19195,1.56431778942e-05,-7.78139e-05 453 | STF1_HUMAN.H10MO.B,5996,-0.000491616993866,0.000106931 454 | HMBX1_HUMAN.H10MO.D,3159,5.54453023227e-05,4.09186e-05 455 | ZN652_HUMAN.H10MO.D,18170,-0.000253309557414,0.00019303 456 | MECP2_HUMAN.H10MO.C,18480,0.000612703695804,0.000184029 457 | VDR_HUMAN.H10MO.B,55597,0.00326347558519,3.94285e-05 458 | ZBT49_HUMAN.H10MO.D,7803,0.000346780173671,9.64105e-05 459 | AP2C_HUMAN.H10MO.A,25206,0.000854711741824,0.000306666 460 | NKX22_HUMAN.H10MO.D,4443,0.000504615271446,7.24792e-05 461 | NFAC3_HUMAN.H10MO.B,11040,0.000231607068643,3.85046e-05 462 | NR2C2_HUMAN.H10MO.A,29741,-0.000225396028574,-0.000122011 463 | VDR_HUMAN.H10MO.S,9073,-0.00103293449419,8.90791e-05 464 | NR2E3_HUMAN.H10MO.C,15499,0.00159860655805,-2.7895e-05 465 | VENTX_HUMAN.H10MO.D,11460,0.00154306092506,0.000178069 466 | SOX15_HUMAN.H10MO.D,9008,0.00027200757203,5.37336e-05 467 | NR1I3_HUMAN.H10MO.C,7395,0.000653825537982,7.37906e-05 468 | CREB5_HUMAN.H10MO.D,7822,-0.000818785921333,-8.46386e-06 469 | RHXF1_HUMAN.H10MO.D,14821,0.0010671938315,0.000173897 470 | TCF7_HUMAN.H10MO.C,10075,-0.00035611506012,1.86563e-05 471 | HOMEZ_HUMAN.H10MO.D,17613,-0.000433065942424,0.000257909 472 | SUH_HUMAN.H10MO.C,10604,0.000738331767789,0.000127375 473 | NR1I3_HUMAN.H10MO.S,3506,0.000114262794133,1.5825e-05 474 | GABP1_HUMAN.H10MO.C,28963,0.000672295298269,0.000145644 475 | FOXI1_HUMAN.H10MO.B,9186,-0.000308052863591,7.43866e-05 476 | KLF4_HUMAN.H10MO.A,50034,0.00156778825954,5.28693e-05 477 | TBX3_HUMAN.H10MO.D,10336,-0.000537540015519,5.34058e-05 478 | GLIS1_HUMAN.H10MO.D,35325,0.00176490090228,0.00021261 479 | DLX5_HUMAN.H10MO.D,13173,0.000609183959146,0.00015533 480 | NKX31_HUMAN.H10MO.C,16903,-0.00043548101907,0.000188798 481 | DLX3_HUMAN.H10MO.C,6969,0.000374272040904,7.53105e-05 482 | TEAD1_HUMAN.H10MO.D,11313,0.000979421480295,7.83503e-05 483 | GATA2_HUMAN.H10MO.A,16927,0.00101580392159,-0.000133067 484 | DLX1_HUMAN.H10MO.D,13348,0.00129994671207,0.000151128 485 | CLOCK_HUMAN.H10MO.D,97157,0.00229037264003,7.06911e-05 486 | ALX1_HUMAN.H10MO.B,10709,0.000664951057018,0.000170022 487 | GLI3_HUMAN.H10MO.B,17564,9.08874777036e-05,0.000387371 488 | FOXG1_HUMAN.H10MO.D,55948,0.000444517700472,0.000401407 489 | HEY1_HUMAN.H10MO.D,16547,-0.000997431970173,0.000276238 490 | ZN219_HUMAN.H10MO.D,83227,0.000536465377511,0.000108272 491 | BHE40_HUMAN.H10MO.A,12111,-0.000252160597993,0.000265747 492 | MEF2B_HUMAN.H10MO.D,7028,-0.000125104748793,-3.72827e-05 493 | HMGA2_HUMAN.H10MO.D,14866,0.000593003912113,-4.96507e-05 494 | IRF2_HUMAN.H10MO.C,41808,-0.000466713457385,0.000153124 495 | MEF2A_HUMAN.H10MO.A,8660,0.000906501703367,4.72665e-05 496 | ZBT7B_HUMAN.H10MO.D,15934,0.000299178521811,-2.25008e-05 497 | HES5_HUMAN.H10MO.D,17867,0.000269796400793,0.000228018 498 | HXA5_HUMAN.H10MO.D,10477,-0.000781940905999,0.000107676 499 | ISX_HUMAN.H10MO.D,11082,-4.05452313127e-05,9.8139e-05 500 | PAX5_HUMAN.H10MO.A,29867,0.00106304761375,2.64645e-05 501 | HES7_HUMAN.H10MO.D,17470,0.000273383362402,0.000165582 502 | HNF4A_HUMAN.H10MO.A,7864,0.000503847762482,-0.000118583 503 | PKNX2_HUMAN.H10MO.D,44415,0.00196394188976,0.000169247 504 | GBX1_HUMAN.H10MO.D,13045,-0.000572512771585,0.000268608 505 | FOXC1_HUMAN.H10MO.C,12451,0.00032048827903,-4.06802e-05 506 | MITF_HUMAN.H10MO.C,9524,0.000493364409926,8.73208e-06 507 | PAX5_HUMAN.H10MO.S,5802,-0.000194939883365,-4.85778e-05 508 | COT1_HUMAN.H10MO.S,8412,0.000361765079773,-9.67979e-05 509 | NFAC2_HUMAN.H10MO.B,16002,0.000977802731398,7.56085e-05 510 | NR2C1_HUMAN.H10MO.C,5879,-0.000987497901769,-2.90275e-05 511 | OTX2_HUMAN.H10MO.C,15143,0.000586478847923,3.41237e-05 512 | MEF2C_HUMAN.H10MO.C,11583,0.000226765206676,0.000110626 513 | NR1I2_HUMAN.H10MO.C,7019,7.32445055652e-05,4.6283e-05 514 | KLF16_HUMAN.H10MO.D,60609,0.000658290203076,-7.08997e-05 515 | ZN589_HUMAN.H10MO.D,20673,0.00113406410398,0.000406623 516 | HXA1_HUMAN.H10MO.C,10505,-0.00160637084268,0.000118732 517 | TBR1_HUMAN.H10MO.D,11337,0.000836684360143,-5.86212e-05 518 | GABPA_HUMAN.H10MO.A,32357,0.00114598003938,0.000225067 519 | ELK4_HUMAN.H10MO.A,16707,0.000109285804062,5.51045e-05 520 | IRF9_HUMAN.H10MO.C,17272,0.00122690520831,0.000294209 521 | FOXO6_HUMAN.H10MO.D,9520,-0.000358577224983,1.86265e-05 522 | DPRX_HUMAN.H10MO.D,11847,0.000171326668659,5.82337e-05 523 | EVI1_HUMAN.H10MO.B,18049,0.000118066187584,-0.000115752 524 | NFKB1_HUMAN.H10MO.B,14907,-0.000139196663499,0.000135273 525 | EHF_HUMAN.H10MO.S,18709,0.000348535707965,-3.24845e-06 526 | HXA2_HUMAN.H10MO.D,8002,0.000408105233642,0.00014329 527 | ZKSC1_HUMAN.H10MO.C,11934,0.00136881132476,-1.48416e-05 528 | CTCF_HUMAN.H10MO.A,26918,0.000978391473036,0.000309616 529 | IKZF1_HUMAN.H10MO.C,15828,-8.17734014954e-05,0.000121832 530 | HEN1_HUMAN.H10MO.C,15264,0.00054962832029,-6.02007e-06 531 | E2F2_HUMAN.H10MO.B,30958,-0.000874241833789,0.000240058 532 | EHF_HUMAN.H10MO.C,98929,0.00742323459301,-0.000491232 533 | PROP1_HUMAN.H10MO.D,12274,0.00115879779838,-3.3319e-05 534 | KLF3_HUMAN.H10MO.D,17733,0.000915437755266,5.07534e-05 535 | SNAI2_HUMAN.H10MO.C,5946,-0.000491039688349,8.16584e-05 536 | TBP_HUMAN.H10MO.C,13482,0.000573179764005,0.00015521 537 | PAX8_HUMAN.H10MO.D,9804,0.000669928335564,-2.16663e-05 538 | GCR_HUMAN.H10MO.S,5496,-0.000430480771104,2.563e-05 539 | ZN350_HUMAN.H10MO.C,19730,0.000490294076608,9.53674e-06 540 | HINFP_HUMAN.H10MO.C,14987,0.00361832101081,0.000115067 541 | HXC12_HUMAN.H10MO.D,13305,-0.000461510360305,9.73046e-05 542 | ZN143_HUMAN.H10MO.A,30879,4.1994477352e-05,6.79195e-05 543 | NR4A3_HUMAN.H10MO.D,6368,0.000276242801252,4.78923e-05 544 | GLIS2_HUMAN.H10MO.D,42572,0.000828415954208,0.000348002 545 | PBX1_HUMAN.H10MO.B,9034,-0.000583856333555,0.000230491 546 | PRGR_HUMAN.H10MO.C,9381,-0.000595042577764,0.000122964 547 | GCR_HUMAN.H10MO.A,10477,0.0007057150989,6.48201e-05 548 | NFIL3_HUMAN.H10MO.C,8098,0.000161361302162,0.000111401 549 | SMAD1_HUMAN.H10MO.D,15242,-6.88494781242e-06,9.11355e-05 550 | HXB7_HUMAN.H10MO.C,11927,0.00030079902589,0.000156224 551 | PAX4_HUMAN.H10MO.D,9759,0.000772724683619,0.000104785 552 | SPIB_HUMAN.H10MO.B,46621,0.000784370833055,6.63102e-05 553 | PPARA_HUMAN.H10MO.S,3080,-0.000133101320497,-3.44515e-05 554 | CEBPA_HUMAN.H10MO.A,6314,0.000421438142054,-5.62966e-05 555 | HXA9_HUMAN.H10MO.D,6767,-0.000552984071746,7.75456e-05 556 | HIC1_HUMAN.H10MO.C,16087,0.000928964639491,0.00020045 557 | ERG_HUMAN.H10MO.B,16169,0.00129845540308,7.04825e-05 558 | PPARA_HUMAN.H10MO.C,20055,-0.00116857882906,-5.37634e-05 559 | FOXO1_HUMAN.H10MO.C,32396,0.00125766896072,0.000198841 560 | GCM1_HUMAN.H10MO.D,13586,0.000475519909983,0.000205934 561 | CUX2_HUMAN.H10MO.D,7588,1.67579761188e-05,7.94828e-05 562 | RORA_HUMAN.H10MO.B,5683,-0.000124235962301,-6.49393e-05 563 | ZBT7A_HUMAN.H10MO.D,33309,0.00039050057458,0.000120491 564 | NR1I2_HUMAN.H10MO.S,3462,0.000217922959293,9.83477e-06 565 | USF1_HUMAN.H10MO.A,10638,0.00026856709996,8.39531e-05 566 | HXC10_HUMAN.H10MO.D,28347,0.000715116234076,0.000437856 567 | SOX18_HUMAN.H10MO.D,14265,-0.000350076750639,-4.02331e-06 568 | LMX1B_HUMAN.H10MO.D,21946,0.00103119179956,0.000218123 569 | CEBPZ_HUMAN.H10MO.D,11164,7.01204699359e-05,0.000150263 570 | ZN410_HUMAN.H10MO.D,12895,-4.92503686864e-05,3.80874e-05 571 | EVX1_HUMAN.H10MO.D,19208,0.00180175374138,0.000216424 572 | FOXB1_HUMAN.H10MO.D,9327,0.0011398031771,5.06043e-05 573 | HXC8_HUMAN.H10MO.D,8503,9.50049857642e-05,8.83043e-05 574 | PAX2_HUMAN.H10MO.S,13159,-0.000202196511139,-0.000126332 575 | SRBP1_HUMAN.H10MO.B,22797,0.00222707776823,-0.000243008 576 | MEIS3_HUMAN.H10MO.D,4727,-8.65850352403e-05,-6.30617e-05 577 | PAX2_HUMAN.H10MO.D,7469,-0.000674629294423,3.56436e-05 578 | FOXP3_HUMAN.H10MO.D,9683,0.000405387118523,4.94719e-05 579 | MSX2_HUMAN.H10MO.D,8800,0.000821689122156,0.000109583 580 | HTF4_HUMAN.H10MO.B,9868,0.000783089863832,5.13792e-05 581 | ARI3A_HUMAN.H10MO.D,39466,0.00244112998708,0.000225246 582 | OLIG2_HUMAN.H10MO.D,9009,8.88782641274e-05,2.16663e-05 583 | CR3L1_HUMAN.H10MO.D,10749,0.000753420913727,7.37011e-05 584 | SP2_HUMAN.H10MO.C,71867,0.00167833560717,-0.000308633 585 | PIT1_HUMAN.H10MO.C,10507,0.000625621363587,-2.74181e-06 586 | TF7L1_HUMAN.H10MO.D,42841,0.000956653380205,0.000245392 587 | FOSL2_HUMAN.H10MO.A,3937,0.000407091508502,1.03116e-05 588 | ZFHX3_HUMAN.H10MO.D,19048,-0.000479250774499,0.000401795 589 | PO3F2_HUMAN.H10MO.D,14339,0.000299096270635,0.000212401 590 | PRD14_HUMAN.H10MO.C,6097,0.000327022244445,5.36144e-05 591 | KLF15_HUMAN.H10MO.D,94112,-0.00109083995777,-0.000181377 592 | SPIC_HUMAN.H10MO.D,83849,0.00372297947505,-4.86374e-05 593 | ZFX_HUMAN.H10MO.C,49225,0.00264761357252,-0.000397116 594 | MNT_HUMAN.H10MO.D,62377,0.000949029844427,-0.000221312 595 | FOXJ3_HUMAN.H10MO.S,35651,0.00254007183246,0.000197411 596 | MTF1_HUMAN.H10MO.C,21062,-0.00031658056137,0.000368446 597 | LHX2_HUMAN.H10MO.D,9312,-0.000245671662649,8.77976e-05 598 | FOXH1_HUMAN.H10MO.A,8275,-4.09141078893e-05,0.000129193 599 | ZEP2_HUMAN.H10MO.D,21088,-0.000731418324604,0.000119001 600 | SOX17_HUMAN.H10MO.D,8791,0.000115312853456,5.00679e-06 601 | FIGLA_HUMAN.H10MO.D,7014,-0.00135774478645,0.000190109 602 | AP2A_HUMAN.H10MO.C,17886,0.00091097261585,0.000256568 603 | FOXM1_HUMAN.H10MO.D,18686,-0.000285823398827,9.91225e-05 604 | PO2F2_HUMAN.H10MO.D,10348,0.00100731038247,1.47223e-05 605 | ETV1_HUMAN.H10MO.B,78101,0.00468174839995,0.000120431 606 | GSC2_HUMAN.H10MO.D,36699,0.00128516277747,-6.23465e-05 607 | HXB2_HUMAN.H10MO.D,3437,0.000697746803555,-1.62125e-05 608 | RUNX2_HUMAN.H10MO.B,9397,-0.000431134471835,-2.57194e-05 609 | FOXO4_HUMAN.H10MO.C,21297,2.76549952888e-05,0.000156909 610 | E2F8_HUMAN.H10MO.D,17254,0.000996528502008,0.000136226 611 | BHA15_HUMAN.H10MO.D,11318,0.000140502221305,6.53565e-05 612 | GCM2_HUMAN.H10MO.D,19915,0.00123417259523,1.88053e-05 613 | ESR2_HUMAN.H10MO.A,11864,-0.000392147582592,-2.41399e-05 614 | NR2F6_HUMAN.H10MO.D,43746,0.00438133349021,-8.04067e-05 615 | HSF2_HUMAN.H10MO.A,4145,-7.10879176161e-05,-2.75671e-05 616 | LBX2_HUMAN.H10MO.D,6621,-0.0014143767687,0.000234663 617 | TEAD4_HUMAN.H10MO.A,8151,-0.000412513157002,9.197e-05 618 | PROX1_HUMAN.H10MO.D,91281,0.00130966360895,-0.000186741 619 | HEY2_HUMAN.H10MO.D,41988,0.00109025056795,0.000113875 620 | ESR2_HUMAN.H10MO.S,7665,0.000383169466784,-9.53674e-07 621 | AHR_HUMAN.H10MO.B,15986,0.00170524723302,0.000223875 622 | KLF1_HUMAN.H10MO.C,47408,0.000807790075231,-0.000116855 623 | ZN639_HUMAN.H10MO.D,50870,0.00445631656259,9.03308e-05 624 | PLAL1_HUMAN.H10MO.D,34389,0.00170167313458,0.000256419 625 | ZEP1_HUMAN.H10MO.D,10209,0.000869241923844,8.63373e-05 626 | BATF_HUMAN.H10MO.S,3489,0.000190704730472,8.67248e-06 627 | DLX2_HUMAN.H10MO.D,8579,-1.11200557354e-05,0.000117242 628 | NF2L2_HUMAN.H10MO.D,5173,-0.000589234991889,4.40478e-05 629 | PRGR_HUMAN.H10MO.S,5496,-0.000393977749844,4.16636e-05 630 | STAT3_HUMAN.H10MO.A,8387,0.00011972305512,0.000117183 631 | HXD12_HUMAN.H10MO.D,10724,-0.000602861693754,-1.16229e-06 632 | OVOL1_HUMAN.H10MO.C,7961,-7.75736793986e-05,8.66354e-05 633 | BATF_HUMAN.H10MO.A,4187,0.00022790121705,-4.78029e-05 634 | MESP1_HUMAN.H10MO.D,10757,-0.000143462999706,8.83639e-05 635 | BARH2_HUMAN.H10MO.D,11046,-0.00151702877324,0.000155389 636 | ZBTB4_HUMAN.H10MO.D,19382,0.000560379925347,-2.69711e-05 637 | HSFY1_HUMAN.H10MO.D,7394,0.000625046724828,0.000119925 638 | TFEB_HUMAN.H10MO.C,7428,-0.000166426498126,-3.51667e-05 639 | HXA13_HUMAN.H10MO.C,14125,-0.00149094046027,0.000116736 640 | ZBTB4_HUMAN.H10MO.S,34513,-0.00109579593784,-9.74834e-05 641 | -------------------------------------------------------------------------------- /feature_importances/analyze_feature_importances.m: -------------------------------------------------------------------------------- 1 | clear; 2 | close all; 3 | addpath('/home/sss1/Desktop/projects/DeepInteractions/feature_importances/brewermap'); 4 | 5 | cell_lines = {'GM12878', 'HeLa-S3', 'HUVEC', 'K562', 'IMR90', 'NHEK'}; 6 | root = '/home/sss1/Desktop/projects/DeepInteractions/feature_importances/SPEID/from_HOCOMOCO_motifs/'; 7 | suffix = '_feature_importance.csv'; 8 | fig_dir = '/home/sss1/Desktop/projects/DeepInteractions/feature_importances/figs/'; 9 | colormap('winter') 10 | 11 | % Load feature importance data for all cell lines 12 | for cell_line_idx = 1:length(cell_lines) 13 | 14 | % enhancers 15 | cell_line = cell_lines{cell_line_idx}; 16 | file_name = [root cell_line '_enhancers' suffix]; 17 | [names, counts, scores, mean_diffs] = read_SPEID_feature_importance(file_name, false); 18 | if strcmp(cell_lines{cell_line_idx}, 'K562') || strcmp(cell_lines{cell_line_idx}, 'NHEK') 19 | mean_diffs = -mean_diffs; 20 | end 21 | if cell_line_idx == 1 22 | importance_enhancers = zeros(length(names), length(cell_lines)); 23 | importance_promoters = zeros(length(names), length(cell_lines)); 24 | count_enhancers = zeros(length(names), length(cell_lines)); 25 | count_promoters = zeros(length(names), length(cell_lines)); 26 | mean_diffs_enhancers = zeros(length(names), length(cell_lines)); 27 | mean_diffs_promoters = zeros(length(names), length(cell_lines)); 28 | end 29 | % Sort features alphabetically by name 30 | [names, I] = sort(names); 31 | importance_enhancers(:, cell_line_idx) = scores(I); 32 | count_enhancers(:, cell_line_idx) = counts(I); 33 | mean_diff_enhancers(:, cell_line_idx) = mean_diffs(I); 34 | 35 | % promoters 36 | file_name = [root cell_line '_promoters' suffix]; 37 | [names, counts, scores, mean_diffs] = read_SPEID_feature_importance(file_name, false); 38 | if strcmp(cell_lines{cell_line_idx}, 'NHEK') 39 | mean_diffs = -mean_diffs; 40 | end 41 | [names, I] = sort(names); 42 | importance_promoters(:, cell_line_idx) = scores(I); 43 | count_promoters(:, cell_line_idx) = counts(I); 44 | mean_diff_promoters(:, cell_line_idx) = mean_diffs(I); 45 | 46 | end 47 | 48 | % Plot feature importance histograms for all cell lines 49 | f = figure; 50 | fig_name = [fig_dir 'importance_hist']; 51 | map = brewermap(3, 'Set1'); 52 | min_importance = min(min(importance_enhancers(:), min(importance_promoters(:)))); 53 | max_importance = max(max(importance_enhancers(:), max(importance_promoters(:)))); 54 | for cell_line_idx = 1:length(cell_lines) 55 | subplot(2, 3, cell_line_idx); 56 | hold all; 57 | h_enhancers = histogram(importance_enhancers(:, cell_line_idx), 'facecolor', map(1, :), 'facealpha', 0.5, 'edgecolor', 'none'); 58 | h_promoters = histogram(importance_promoters(:, cell_line_idx), 'facecolor', map(2, :), 'facealpha', 0.5, 'edgecolor', 'none'); 59 | BinWidth = mean([h_enhancers.BinWidth h_promoters.BinWidth]); 60 | h_promoters.BinWidth = BinWidth; 61 | h_enhancers.BinWidth = BinWidth; 62 | xlim([min_importance max_importance]); 63 | if cell_line_idx == 3 64 | legend({'Enhancers', 'Promoters'}, 'FontSize', 20); 65 | end 66 | title(cell_lines{cell_line_idx}, 'FontSize', 24); 67 | if cell_line_idx > 3 % Bottom row 68 | xlabel('Feature Importance', 'FontSize', 20); 69 | end 70 | if mod(cell_line_idx, 3) == 1 % Left column 71 | ylabel('Frequency', 'FontSize', 20); 72 | end 73 | set(gca, 'FontSize', 14); 74 | end 75 | set(gcf, 'Position', get(0,'Screensize')); % Maximize figure 76 | saveas(f, [fig_name '.fig']); 77 | saveas(f, [fig_name '.png']); 78 | set(f, 'Units', 'Inches'); pos = get(f, 'Position'); set(f, 'PaperPositionMode', 'Auto', 'PaperUnits', 'Inches', 'PaperSize', [pos(3), pos(4)]) 79 | print(f, [fig_name '.pdf'], '-dpdf', '-r0') 80 | 81 | % Plot feature mean_diff histograms for all cell lines 82 | f = figure; 83 | fig_name = [fig_dir 'mean_diff_hist']; 84 | map = brewermap(3, 'Set1'); 85 | min_mean_diff = min(min(mean_diff_enhancers(:), min(mean_diff_promoters(:)))); 86 | max_mean_diff = max(max(mean_diff_enhancers(:), max(mean_diff_promoters(:)))); 87 | for cell_line_idx = 1:length(cell_lines) 88 | subplot(2, 3, cell_line_idx); 89 | hold all; 90 | h_enhancers = histogram(mean_diff_enhancers(:, cell_line_idx), 'facecolor', map(1, :), 'facealpha', 0.5, 'edgecolor', 'none'); 91 | h_promoters = histogram(mean_diff_promoters(:, cell_line_idx), 'facecolor', map(2, :), 'facealpha', 0.5, 'edgecolor', 'none'); 92 | BinWidth = mean([h_enhancers.BinWidth h_promoters.BinWidth]); 93 | h_promoters.BinWidth = BinWidth; 94 | h_enhancers.BinWidth = BinWidth; 95 | xlim([min_mean_diff max_mean_diff]); 96 | if cell_line_idx == 3 97 | legend({'Enhancers', 'Promoters'}, 'FontSize', 20); 98 | end 99 | title(cell_lines{cell_line_idx}, 'FontSize', 24); 100 | if cell_line_idx > 3 % Bottom row 101 | xlabel('Feature Repression Factor', 'FontSize', 20); 102 | end 103 | if mod(cell_line_idx, 3) == 1 % Left column 104 | ylabel('Frequency', 'FontSize', 20); 105 | end 106 | set(gca, 'FontSize', 14); 107 | end 108 | set(gcf, 'Position', get(0,'Screensize')); % Maximize figure 109 | saveas(f, [fig_name '.fig']); 110 | saveas(f, [fig_name '.png']); 111 | set(f, 'Units', 'Inches'); pos = get(f, 'Position'); set(f, 'PaperPositionMode', 'Auto', 'PaperUnits', 'Inches', 'PaperSize', [pos(3), pos(4)]) 112 | print(f, [fig_name '.pdf'], '-dpdf', '-r0') 113 | 114 | % Plot feature importance over feature counts for all cell lines 115 | f = figure; 116 | fig_name = [fig_dir 'importance_over_count']; 117 | for cell_line_idx = 1:length(cell_lines) 118 | subplot(2, 3, cell_line_idx); 119 | hold all; 120 | scatter(count_enhancers(:, cell_line_idx), importance_enhancers(:, cell_line_idx)); 121 | scatter(count_promoters(:, cell_line_idx), importance_promoters(:, cell_line_idx)); 122 | set(gca, 'xscale', 'log'); 123 | if cell_line_idx == 3 124 | legend({'Enhancers', 'Promoters'}, 'FontSize', 20); 125 | end 126 | title(cell_lines{cell_line_idx}, 'FontSize', 24); 127 | if cell_line_idx > 3 % Bottom row 128 | xlabel('Motif Count', 'FontSize', 20); 129 | end 130 | if mod(cell_line_idx, 3) == 1 % Left column 131 | ylabel('Motif Importance', 'FontSize', 20); 132 | end 133 | set(gca, 'FontSize', 14); 134 | end 135 | set(gcf, 'Position', get(0,'Screensize')); % Maximize figure 136 | saveas(f, [fig_name '.fig']); 137 | saveas(f, [fig_name '.png']); 138 | set(f, 'Units', 'Inches'); pos = get(f, 'Position'); set(f, 'PaperPositionMode', 'Auto', 'PaperUnits', 'Inches', 'PaperSize', [pos(3), pos(4)]) 139 | print(f, [fig_name '.pdf'], '-dpdf', '-r0') 140 | 141 | % Dispay all feature importances as image 142 | f = figure; 143 | fig_name = [fig_dir 'importance_all']; 144 | num_top_features = 639; 145 | subplot(1, 2, 1); 146 | ranked = tiedrank(-mean_diff_enhancers); 147 | [sorted, I] = sort(mean(ranked, 2)); 148 | imagesc(ranked(I(1:num_top_features), :)); 149 | set(gca, 'XTick', 1:6); set(gca, 'XTickLabel', cell_lines); set(gca,'XTickLabelRotation',45); 150 | title('Enhancers', 'FontSize', 24); 151 | set(gca, 'FontSize', 14); 152 | subplot(1, 2, 2); 153 | ranked = tiedrank(-mean_diff_promoters); 154 | [sorted, I] = sort(mean(ranked, 2)); 155 | imagesc(ranked(I(1:num_top_features), :)); 156 | set(gca, 'XTick', 1:6); set(gca, 'XTickLabel', cell_lines); set(gca,'XTickLabelRotation',45); 157 | title('Promoters', 'FontSize', 24); 158 | set(gca, 'FontSize', 14); 159 | c = colorbar; set(c, 'YDir', 'reverse' ); ylabel(c, 'Feature Rank', 'FontSize', 20); 160 | set(gcf, 'Position', get(0,'Screensize')); % Maximize figure 161 | saveas(f, [fig_name '.fig']); 162 | saveas(f, [fig_name '.png']); 163 | set(f, 'Units', 'Inches'); pos = get(f, 'Position'); set(f, 'PaperPositionMode', 'Auto', 'PaperUnits', 'Inches', 'PaperSize', [pos(3), pos(4)]) 164 | print(f, [fig_name '.pdf'], '-dpdf', '-r0') 165 | 166 | % Display feature importances of mean-rank-top-20 features for all cell lines as image 167 | f = figure; 168 | fig_name = [fig_dir 'importance_top_20']; 169 | num_top_features = 20; 170 | subplot(1, 2, 1); 171 | ranked = tiedrank(-importance_enhancers); 172 | [sorted, I] = sort(mean(ranked, 2)); 173 | imagesc(ranked(I(1:num_top_features), :), [min(ranked(:)) max(ranked(:))]); 174 | set(gca, 'XTick', 1:6); set(gca, 'XTickLabel', cell_lines); set(gca,'XTickLabelRotation',45); 175 | set(gca, 'YTick', 1:num_top_features); set(gca, 'YTickLabel', names(I(1:num_top_features))); 176 | title('Enhancers', 'FontSize', 24); 177 | set(gca, 'FontSize', 14); 178 | subplot(1, 2, 2); 179 | ranked = tiedrank(-importance_promoters); 180 | [sorted, I] = sort(mean(ranked, 2)); 181 | imagesc(ranked(I(1:num_top_features), :), [min(ranked(:)) max(ranked(:))]); 182 | set(gca, 'XTick', 1:6); set(gca, 'XTickLabel', cell_lines); set(gca,'XTickLabelRotation',45); 183 | set(gca, 'YTick', 1:num_top_features); set(gca, 'YTickLabel', names(I(1:num_top_features))); 184 | title('Promoters', 'FontSize', 24); 185 | set(gca, 'FontSize', 14); 186 | c = colorbar; set(c, 'YDir', 'reverse' ); ylabel(c, 'Feature Rank', 'FontSize', 20); 187 | set(gcf, 'Position', get(0,'Screensize')); % Maximize figure 188 | saveas(f, [fig_name '.fig']); 189 | saveas(f, [fig_name '.png']); 190 | set(f, 'Units', 'Inches'); pos = get(f, 'Position'); set(f, 'PaperPositionMode', 'Auto', 'PaperUnits', 'Inches', 'PaperSize', [pos(3), pos(4)]) 191 | print(f, [fig_name '.pdf'], '-dpdf', '-r0') 192 | 193 | % Plot feature mean_diff over feature counts for all cell lines 194 | f = figure; 195 | fig_name = [fig_dir 'mean_diff_over_count']; 196 | for cell_line_idx = 1:length(cell_lines) 197 | subplot(2, 3, cell_line_idx); 198 | hold all; 199 | scatter(count_enhancers(:, cell_line_idx), mean_diff_enhancers(:, cell_line_idx)); 200 | scatter(count_promoters(:, cell_line_idx), mean_diff_promoters(:, cell_line_idx)); 201 | set(gca, 'xscale', 'log'); 202 | if cell_line_idx == 3 203 | legend({'Enhancers', 'Promoters'}, 'FontSize', 20); 204 | end 205 | title(cell_lines{cell_line_idx}, 'FontSize', 24); 206 | if cell_line_idx > 3 % Bottom row 207 | xlabel('Motif Count', 'FontSize', 20); 208 | end 209 | if mod(cell_line_idx, 3) == 1 % Left column 210 | ylabel('Motif Importance', 'FontSize', 20); 211 | end 212 | set(gca, 'FontSize', 14); 213 | end 214 | set(gcf, 'Position', get(0,'Screensize')); % Maximize figure 215 | saveas(f, [fig_name '.fig']); 216 | saveas(f, [fig_name '.png']); 217 | set(f, 'Units', 'Inches'); pos = get(f, 'Position'); set(f, 'PaperPositionMode', 'Auto', 'PaperUnits', 'Inches', 'PaperSize', [pos(3), pos(4)]) 218 | print(f, [fig_name '.pdf'], '-dpdf', '-r0') 219 | 220 | % Display all feature mean_diffs as image 221 | f = figure; 222 | fig_name = [fig_dir 'mean_diff_all']; 223 | num_top_features = 639; 224 | subplot(1, 2, 1); 225 | ranked = tiedrank(-mean_diff_enhancers); 226 | [sorted, I] = sort(mean(ranked, 2)); 227 | imagesc(ranked(I(1:num_top_features), :)); 228 | set(gca, 'XTick', 1:6); set(gca, 'XTickLabel', cell_lines); set(gca,'XTickLabelRotation',45); 229 | title('Enhancers', 'FontSize', 24); 230 | set(gca, 'FontSize', 14); 231 | subplot(1, 2, 2); 232 | ranked = tiedrank(-mean_diff_promoters); 233 | [sorted, I] = sort(mean(ranked, 2)); 234 | imagesc(ranked(I(1:num_top_features), :)); 235 | set(gca, 'XTick', 1:6); set(gca, 'XTickLabel', cell_lines); set(gca,'XTickLabelRotation',45); 236 | title('Promoters', 'FontSize', 24); 237 | set(gca, 'FontSize', 14); 238 | c = colorbar; set(c, 'YDir', 'reverse' ); ylabel(c, 'Feature Rank', 'FontSize', 20); 239 | set(gcf, 'Position', get(0,'Screensize')); % Maximize figure 240 | saveas(f, [fig_name '.fig']); 241 | saveas(f, [fig_name '.png']); 242 | set(f, 'Units', 'Inches'); pos = get(f, 'Position'); set(f, 'PaperPositionMode', 'Auto', 'PaperUnits', 'Inches', 'PaperSize', [pos(3), pos(4)]) 243 | print(f, [fig_name '.pdf'], '-dpdf', '-r0') 244 | 245 | % Display feature mean_diffs of mean-rank-top-20 features for all cell lines as image 246 | f = figure; 247 | fig_name = [fig_dir 'mean_diff_top_20']; 248 | num_top_features = 20; 249 | subplot(1, 2, 1); 250 | ranked = tiedrank(-mean_diff_enhancers); 251 | [sorted, I] = sort(mean(ranked, 2)); 252 | imagesc(ranked(I(1:num_top_features), :), [min(ranked(:)) max(ranked(:))]); 253 | set(gca, 'XTick', 1:6); set(gca, 'XTickLabel', cell_lines); set(gca,'XTickLabelRotation',45); 254 | set(gca, 'YTick', 1:num_top_features); set(gca, 'YTickLabel', names(I(1:num_top_features))); 255 | title('Enhancers', 'FontSize', 24); 256 | set(gca, 'FontSize', 14); 257 | subplot(1, 2, 2); 258 | ranked = tiedrank(-mean_diff_promoters); 259 | [sorted, I] = sort(mean(ranked, 2)); 260 | imagesc(ranked(I(1:num_top_features), :), [min(ranked(:)) max(ranked(:))]); 261 | set(gca, 'XTick', 1:6); set(gca, 'XTickLabel', cell_lines); set(gca,'XTickLabelRotation',45); 262 | set(gca, 'YTick', 1:num_top_features); set(gca, 'YTickLabel', names(I(1:num_top_features))); 263 | title('Promoters', 'FontSize', 24); 264 | set(gca, 'FontSize', 14); 265 | c = colorbar; set(c, 'YDir', 'reverse' ); ylabel(c, 'Feature Rank', 'FontSize', 20); 266 | set(gcf, 'Position', get(0,'Screensize')); % Maximize figure 267 | saveas(f, [fig_name '.fig']); 268 | saveas(f, [fig_name '.png']); 269 | set(f, 'Units', 'Inches'); pos = get(f, 'Position'); set(f, 'PaperPositionMode', 'Auto', 'PaperUnits', 'Inches', 'PaperSize', [pos(3), pos(4)]) 270 | print(f, [fig_name '.pdf'], '-dpdf', '-r0') 271 | -------------------------------------------------------------------------------- /feature_importances/collect_SPEID_results.m: -------------------------------------------------------------------------------- 1 | cell_lines = {'GM12878', 'HeLa-S3', 'HUVEC', 'IMR90', 'K562', 'NHEK'}; 2 | 3 | root = '/home/sss1/Desktop/projects/DeepInteractions/feature_importances/SPEID/from_HOCOMOCO_motifs/'; 4 | suffix = '_feature_importance.csv'; 5 | 6 | for cell_line_idx = 1:length(cell_lines) 7 | 8 | cell_line = cell_lines{cell_line_idx}; 9 | 10 | % Load enhancer results 11 | file_name = [root cell_line '_enhancers' suffix]; 12 | disp(['Reading file ' file_name]); 13 | [names, ~, scores] = read_SPEID_feature_importance(file_name, false); 14 | [names, I] = sort(names); 15 | scores = scores(I); 16 | if cell_line_idx == 1 % initialize output matrix, now that we know # features 17 | importance_enhancers = zeros(length(names), length(cell_lines)); 18 | importance_promoters = zeros(length(names), length(cell_lines)); 19 | end 20 | importance_enhancers(:, cell_line_idx) = scores; 21 | 22 | % Load promoter results 23 | file_name = [root cell_line '_promoters' suffix]; 24 | disp(['Reading file ' file_name]); 25 | [names, ~, scores] = read_SPEID_feature_importance(file_name, false); 26 | [names, I] = sort(names); 27 | scores = scores(I); 28 | importance_promoters(:, cell_line_idx) = scores; 29 | 30 | end 31 | 32 | save([root 'collected_results.mat'], ... 33 | 'cell_lines', ... 34 | 'names', ... 35 | 'importance_enhancers', ... 36 | 'importance_promoters'); 37 | -------------------------------------------------------------------------------- /feature_importances/name_scatter_plot.m: -------------------------------------------------------------------------------- 1 | function scatter_names(x, y, names) 2 | 3 | scatter(x, y); 4 | 5 | % displacement so text does not overlay the data points 6 | dx = (max(x) - min(x))/100; 7 | dy = (max(y) - min(y))/100; 8 | 9 | text(x+dx, y+dy, names); 10 | 11 | end 12 | -------------------------------------------------------------------------------- /feature_importances/read_SPEID_feature_importance.m: -------------------------------------------------------------------------------- 1 | function [names, counts, scores, mean_diffs] = read_SPEID_feature_importance(fileName, to_sort) 2 | 3 | if nargin < 2 % by default, sort results by feature importance 4 | to_sort = true; 5 | end 6 | 7 | fileID = fopen(fileName); 8 | C = textscan(fileID, '%s %f %f %f', 'Delimiter', ',', 'headerLines', 1); 9 | fclose(fileID); 10 | 11 | % extract columns by name 12 | names = C{1}; 13 | counts = C{2}; 14 | scores = C{3}; 15 | mean_diffs = C{4}; 16 | 17 | if to_sort 18 | % sort motifs by score (descending) 19 | [scores, I] = sort(scores, 'descend'); 20 | names = names(I); 21 | counts = counts(I); 22 | mean_diffs = mean_diffs(I); 23 | end 24 | 25 | end 26 | -------------------------------------------------------------------------------- /figs/SPEID_versus_TargetFinder.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ma-compbio/SPEID/edb9d9bffab131add3fad0f58f969f7cc3654100/figs/SPEID_versus_TargetFinder.pdf -------------------------------------------------------------------------------- /figs/importance_all.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ma-compbio/SPEID/edb9d9bffab131add3fad0f58f969f7cc3654100/figs/importance_all.pdf -------------------------------------------------------------------------------- /figs/importance_hist.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ma-compbio/SPEID/edb9d9bffab131add3fad0f58f969f7cc3654100/figs/importance_hist.pdf -------------------------------------------------------------------------------- /figs/importance_over_count.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ma-compbio/SPEID/edb9d9bffab131add3fad0f58f969f7cc3654100/figs/importance_over_count.pdf -------------------------------------------------------------------------------- /figs/importance_top_20.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ma-compbio/SPEID/edb9d9bffab131add3fad0f58f969f7cc3654100/figs/importance_top_20.pdf -------------------------------------------------------------------------------- /figs/mean_diff_all.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ma-compbio/SPEID/edb9d9bffab131add3fad0f58f969f7cc3654100/figs/mean_diff_all.pdf -------------------------------------------------------------------------------- /figs/mean_diff_hist.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ma-compbio/SPEID/edb9d9bffab131add3fad0f58f969f7cc3654100/figs/mean_diff_hist.pdf -------------------------------------------------------------------------------- /figs/mean_diff_over_count.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ma-compbio/SPEID/edb9d9bffab131add3fad0f58f969f7cc3654100/figs/mean_diff_over_count.pdf -------------------------------------------------------------------------------- /figs/mean_diff_top_20.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ma-compbio/SPEID/edb9d9bffab131add3fad0f58f969f7cc3654100/figs/mean_diff_top_20.pdf -------------------------------------------------------------------------------- /pairwise/basic_training.py: -------------------------------------------------------------------------------- 1 | # Basic python and data processing imports 2 | import numpy as np 3 | np.set_printoptions(suppress=True) # Suppress scientific notation when printing small 4 | import h5py 5 | 6 | import load_data_pairs as ld # my scripts for loading data 7 | import build_small_model as bm # Keras specification of SPEID model 8 | 9 | # import matplotlib.pyplot as plt 10 | from datetime import datetime 11 | import util 12 | 13 | # Keras imports 14 | from keras.optimizers import RMSprop, Adam 15 | from keras.callbacks import ModelCheckpoint, EarlyStopping, Callback 16 | 17 | cell_lines = ['GM12878', 'HeLa-S3', 'HUVEC', 'IMR90', 'K562', 'NHEK'] 18 | 19 | # Model training parameters 20 | num_epochs = 32 21 | batch_size = 100 22 | training_frac = 0.9 # fraction of data to use for training 23 | 24 | t = datetime.now().strftime('%Y-%m-%d-%H:%M:%S') 25 | opt = Adam(lr = 1e-5) # opt = RMSprop(lr = 1e-6) 26 | 27 | data_path = '/home/sss1/Desktop/projects/DeepInteractions/data/uniform_len/original/all_data.h5' 28 | 29 | for cell_line in cell_lines: 30 | print 'Loading ' + cell_line + ' data from ' + data_path 31 | X_enhancers = None 32 | X_promoters = None 33 | labels = None 34 | with h5py.File(data_path, 'r') as hf: 35 | X_enhancers = np.array(hf.get(cell_line + '_X_enhancers')).transpose((0, 2, 1)) 36 | X_promoters = np.array(hf.get(cell_line + '_X_promoters')).transpose((0, 2, 1)) 37 | labels = np.array(hf.get(cell_line + 'labels')) 38 | 39 | model = bm.build_model(use_JASPAR = False) 40 | 41 | model.compile(loss = 'binary_crossentropy', 42 | optimizer = opt, 43 | metrics = ["accuracy"]) 44 | 45 | model.summary() 46 | 47 | 48 | # Define custom callback that prints/plots performance at end of each epoch 49 | class ConfusionMatrix(Callback): 50 | def on_train_begin(self, logs = {}): 51 | self.epoch = 0 52 | self.precisions = [] 53 | self.recalls = [] 54 | self.f1_scores = [] 55 | self.losses = [] 56 | self.training_losses = [] 57 | self.training_accs = [] 58 | self.accs = [] 59 | plt.ion() 60 | 61 | def on_epoch_end(self, batch, logs = {}): 62 | self.training_losses.append(logs.get('loss')) 63 | self.training_accs.append(logs.get('acc')) 64 | self.epoch += 1 65 | val_predict = model.predict_classes([X_enhancers, X_promoters], batch_size = batch_size, verbose = 0) 66 | util.print_live(self, labels, val_predict, logs) 67 | if self.epoch > 1: # need at least two time points to plot 68 | util.plot_live(self) 69 | 70 | # print '\nlabels.mean(): ' + str(labels.mean()) 71 | print 'Data sizes: ' 72 | print '[X_enhancers, X_promoters]: [' + str(np.shape(X_enhancers)) + ', ' + str(np.shape(X_promoters)) + ']' 73 | print 'labels: ' + str(np.shape(labels)) 74 | 75 | # Instantiate callbacks 76 | confusionMatrix = ConfusionMatrix() 77 | checkpoint_path = "/home/sss1/Desktop/projects/DeepInteractions/weights/test-delete-this-" + cell_line + "-basic-" + t + ".hdf5" 78 | checkpointer = ModelCheckpoint(filepath=checkpoint_path, verbose = 1) 79 | 80 | print 'Running fully trainable model for exactly ' + str(num_epochs) + ' epochs...' 81 | model.fit([X_enhancers, X_promoters], 82 | [labels], 83 | # validation_data = ([X_enhancer, X_promoter], y_val), 84 | batch_size = batch_size, 85 | nb_epoch = num_epochs, 86 | shuffle = True, 87 | callbacks=[confusionMatrix, checkpointer] 88 | ) 89 | -------------------------------------------------------------------------------- /pairwise/build_small_model.py: -------------------------------------------------------------------------------- 1 | import util 2 | 3 | # Keras imports 4 | from keras.layers import Input, Convolution1D, MaxPooling1D, Merge, Dropout, Flatten, Dense, BatchNormalization, LSTM, Activation, Bidirectional 5 | from keras.optimizers import RMSprop, Adam 6 | from keras.callbacks import ModelCheckpoint, EarlyStopping, Callback 7 | from keras.models import Sequential 8 | from keras.regularizers import l1, l2 9 | 10 | # model parameters 11 | enhancer_length = 3000 # TODO: get this from input 12 | promoter_length = 2000 # TODO: get this from input 13 | n_kernels = 200 # Number of kernels; used to be 1024 14 | filter_length = 40 # Length of each kernel 15 | LSTM_out_dim = 50 # Output direction of ONE DIRECTION of LSTM; used to be 512 16 | dense_layer_size = 800 17 | 18 | # Convolutional/maxpooling layers to extract prominent motifs 19 | # Separate identically initialized convolutional layers are trained for 20 | # enhancers and promoters 21 | # Define enhancer layers 22 | enhancer_conv_layer = Convolution1D(input_dim = 4, 23 | input_length = enhancer_length, 24 | nb_filter = n_kernels, 25 | filter_length = filter_length, 26 | border_mode = "valid", 27 | subsample_length = 1, 28 | W_regularizer = l2(1e-5)) 29 | enhancer_max_pool_layer = MaxPooling1D(pool_length = filter_length/2, stride = filter_length/2) 30 | 31 | # Build enhancer branch 32 | enhancer_branch = Sequential() 33 | enhancer_branch.add(enhancer_conv_layer) 34 | enhancer_branch.add(Activation("relu")) 35 | enhancer_branch.add(enhancer_max_pool_layer) 36 | 37 | # Define promoter layers branch: 38 | promoter_conv_layer = Convolution1D(input_dim = 4, 39 | input_length = promoter_length, 40 | nb_filter = n_kernels, 41 | filter_length = filter_length, 42 | border_mode = "valid", 43 | subsample_length = 1, 44 | W_regularizer = l2(1e-5)) 45 | promoter_max_pool_layer = MaxPooling1D(pool_length = filter_length/2, stride = filter_length/2) 46 | 47 | # Build promoter branch 48 | promoter_branch = Sequential() 49 | promoter_branch.add(promoter_conv_layer) 50 | promoter_branch.add(Activation("relu")) 51 | promoter_branch.add(promoter_max_pool_layer) 52 | 53 | # Define main model layers 54 | # Concatenate outputs of enhancer and promoter convolutional layers 55 | merge_layer = Merge([enhancer_branch, promoter_branch], 56 | mode = 'concat', 57 | concat_axis = 1) 58 | 59 | 60 | # Bidirectional LSTM to extract combinations of motifs 61 | biLSTM_layer = Bidirectional(LSTM(input_dim = n_kernels, 62 | output_dim = LSTM_out_dim, 63 | return_sequences = True)) 64 | 65 | # Dense layer to allow nonlinearities 66 | dense_layer = Dense(output_dim = dense_layer_size, 67 | init = "glorot_uniform", 68 | W_regularizer = l2(1e-6)) 69 | 70 | # Logistic regression layer to make final binary prediction 71 | LR_classifier_layer = Dense(output_dim = 1) 72 | 73 | 74 | def build_model(use_JASPAR = True): 75 | 76 | # A single downstream model merges the enhancer and promoter branches 77 | # Build main (merged) branch 78 | # Using batch normalization seems to inhibit retraining, probably because the 79 | # point of retraining is to learn (external) covariate shift 80 | model = Sequential() 81 | model.add(merge_layer) 82 | model.add(BatchNormalization()) 83 | model.add(Dropout(0.25)) 84 | model.add(biLSTM_layer) 85 | model.add(BatchNormalization()) 86 | model.add(Dropout(0.5)) 87 | model.add(Flatten()) 88 | model.add(dense_layer) 89 | model.add(BatchNormalization()) 90 | model.add(Activation("relu")) 91 | model.add(Dropout(0.5)) 92 | model.add(LR_classifier_layer) 93 | model.add(BatchNormalization()) 94 | model.add(Activation("sigmoid")) 95 | 96 | # Read in and initialize convolutional layers with motifs from JASPAR 97 | if use_JASPAR: 98 | util.initialize_with_JASPAR(enhancer_conv_layer, promoter_conv_layer) 99 | 100 | return model 101 | 102 | def build_frozen_model(): 103 | 104 | # Freeze all but the dense layers of the network 105 | enhancer_conv_layer.trainable = False 106 | enhancer_max_pool_layer.trainable = False 107 | promoter_conv_layer.trainable = False 108 | promoter_max_pool_layer.trainable = False 109 | biLSTM_layer.trainable = False 110 | 111 | # TODO: Figure out how to remove layers after loading weights 112 | 113 | model = Sequential() 114 | model.add(merge_layer) 115 | model.add(BatchNormalization()) 116 | model.add(Dropout(0.25)) 117 | model.add(biLSTM_layer) 118 | model.add(BatchNormalization()) 119 | model.add(Dropout(0.5)) 120 | model.add(Flatten()) 121 | model.add(dense_layer) 122 | model.add(BatchNormalization()) 123 | model.add(Activation("relu")) 124 | model.add(Dropout(0.5)) 125 | model.add(LR_classifier_layer) 126 | model.add(BatchNormalization()) 127 | model.add(Activation("sigmoid")) 128 | 129 | return model 130 | -------------------------------------------------------------------------------- /pairwise/data_processing/combine_hdf5s.py: -------------------------------------------------------------------------------- 1 | # Basic python and data processing imports 2 | import numpy as np 3 | import h5py 4 | from sklearn.utils import shuffle 5 | import load_data_pairs as ld # my own scripts for loading data 6 | 7 | # input data paths 8 | combined_name = '4lines' 9 | data_root = '/home/sss1/Desktop/projects/DeepInteractions/data/uniform_len/' 10 | 11 | # output data path 12 | out_path = data_root + combined_name + '/' + combined_name + '_ep_split.h5' 13 | 14 | source_prefixes = ['GM12878', 'HeLa-S3', 'IMR90', 'K562'] 15 | 16 | # Sizes of fixed data dimensions 17 | enhancers_length = 3000 18 | promoters_length = 2000 19 | num_bases = 4 20 | 21 | X_enhancers_train_combined = np.zeros((0, enhancers_length, num_bases)) 22 | X_promoters_train_combined = np.zeros((0, promoters_length, num_bases)) 23 | y_train_combined = np.zeros((0,)) 24 | X_enhancers_train_aug_combined = np.zeros((0, enhancers_length, num_bases)) 25 | X_promoters_train_aug_combined = np.zeros((0, promoters_length, num_bases)) 26 | y_train_aug_combined = np.zeros((0,)) 27 | 28 | for prefix in source_prefixes: 29 | 30 | source_path = data_root + prefix + '/' + prefix + '_ep_split.h5' 31 | print 'Loading ' + prefix + ' data...' 32 | X_enhancers_train, X_promoters_train, y_train = ld.load_hdf5_ep_split(source_path) 33 | X_enhancers_train_aug, X_promoters_train_aug, y_train_aug = ld.load_hdf5_ep_split_aug(source_path) 34 | 35 | print 'Concatenating ' + prefix + ' data...' 36 | X_enhancers_train_combined = np.concatenate((X_enhancers_train_combined, X_enhancers_train)) 37 | X_promoters_train_combined = np.concatenate((X_promoters_train_combined, X_promoters_train)) 38 | y_train_combined = np.concatenate((y_train_combined, y_train)) 39 | X_enhancers_train_aug_combined = np.concatenate((X_enhancers_train_aug_combined, X_enhancers_train_aug)) 40 | X_promoters_train_aug_combined = np.concatenate((X_promoters_train_aug_combined, X_promoters_train_aug)) 41 | y_train_aug_combined = np.concatenate((y_train_aug_combined, y_train_aug)) 42 | 43 | # Save state of random number generator so we can jointly shuffle data 44 | print 'Shuffling data...' 45 | rng_state = np.random.get_state() 46 | np.random.set_state(rng_state) 47 | np.random.shuffle(X_enhancers_train_combined) 48 | np.random.set_state(rng_state) 49 | np.random.shuffle(y_train_combined) 50 | np.random.set_state(rng_state) 51 | np.random.shuffle(X_promoters_train_combined) 52 | np.random.set_state(rng_state) 53 | np.random.shuffle(X_enhancers_train_aug_combined) 54 | np.random.set_state(rng_state) 55 | np.random.shuffle(y_train_aug_combined) 56 | np.random.set_state(rng_state) 57 | np.random.shuffle(X_promoters_train_aug_combined) 58 | 59 | print 'Writing data...' 60 | with h5py.File(out_path, 'w') as hf: 61 | hf.create_dataset('X_enhancers_train', data = X_enhancers_train) 62 | hf.create_dataset('y_train', data = y_train) 63 | hf.create_dataset('X_promoters_train', data = X_promoters_train) 64 | hf.create_dataset('X_enhancers_train_aug', data = X_enhancers_train_aug) 65 | hf.create_dataset('y_train_aug', data = y_train_aug) 66 | hf.create_dataset('X_promoters_train_aug', data = X_promoters_train_aug) 67 | -------------------------------------------------------------------------------- /pairwise/data_processing/load_data_pairs.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | import sys 3 | import h5py 4 | 5 | def load_hdf5(path): 6 | with h5py.File(path,'r') as hf: 7 | X_train = np.array(hf.get('X_train')) 8 | y_train = np.array(hf.get('y_train')) 9 | return X_train, y_train 10 | 11 | def load_hdf5_ep_split(path): 12 | with h5py.File(path,'r') as hf: 13 | X_enhancers_train = np.array(hf.get('X_enhancers_train')) 14 | X_promoters_train = np.array(hf.get('X_promoters_train')) 15 | y_train = np.array(hf.get('y_train')) 16 | return X_enhancers_train, X_promoters_train, y_train 17 | 18 | def load_hdf5_ep_split_aug(path): 19 | with h5py.File(path,'r') as hf: 20 | X_enhancers_train = np.array(hf.get('X_enhancers_train_aug')) 21 | X_promoters_train = np.array(hf.get('X_promoters_train_aug')) 22 | y_train = np.array(hf.get('y_train_aug')) 23 | return X_enhancers_train, X_promoters_train, y_train 24 | 25 | # Reads the specified files and builds num_samples X 4 X seg_length arrays for 26 | # both positive and negative versions of the enhancers and promoter data sets 27 | # Outputs: 28 | # 1) concatenations of enhancer and promoter segments, for all positive and 29 | # negative samples 30 | # 2) labels (0 for negative, 1 for positive) for corresponding samples 31 | def load_full_data(pos_enhancer_f_name, pos_promoter_f_name, neg_enhancer_f_name, neg_promoter_f_name): 32 | 33 | print 'Loading positive samples...' 34 | positive = load_ep_pairs(pos_enhancer_f_name, pos_promoter_f_name) 35 | positive_labels = np.ones(positive.shape[0]) 36 | 37 | print 'Loading negative samples...' 38 | negative = load_ep_pairs(neg_enhancer_f_name, neg_promoter_f_name) 39 | negative_labels = np.zeros(negative.shape[0]) 40 | 41 | samples = np.concatenate((positive, negative), 0) 42 | labels = np.concatenate((positive_labels, negative_labels), 0) 43 | return samples, labels 44 | 45 | # TODO!! 46 | def load_imbalanced_data(pos_neg_ratio, pos_enhancer_f_name, pos_promoter_f_name, neg_enhancer_f_name, neg_promoter_f_name): 47 | 48 | print 'Loading positive samples...' 49 | positive = load_ep_pairs(pos_enhancer_f_name, pos_promoter_f_name) 50 | n_pos = positive.shape[0] # original number of positive samples 51 | n_pos_sub = round(n_pos * pos_neg_ratio) # number of subsampled positive samples 52 | 53 | positive_subsampled = positive[np.random.choice(n_pos, n_pos_sub, replace=False),:, :] 54 | positive_labels = np.ones(n_pos_sub) 55 | 56 | print 'Loading negative samples...' 57 | negative = load_ep_pairs(neg_enhancer_f_name, neg_promoter_f_name) 58 | negative_labels = np.zeros(negative.shape[0]) 59 | 60 | samples = np.concatenate((positive_subsampled, negative), 0) 61 | labels = np.concatenate((positive_labels, negative_labels), 0) 62 | return samples, labels 63 | 64 | # Reads the specified files and builds num_samples X 4 X seg_length arrays for 65 | # both the enhancers and promoter data sets; returns concatenations of enhancer 66 | # and promoter segments 67 | def load_ep_pairs(enhancer_f_name, promoter_f_name): 68 | 69 | print 'Loading enhancer data from ' + enhancer_f_name + '...' 70 | enhancers = load_file(enhancer_f_name) 71 | 72 | # # Code to just load enhancers 73 | # print 'Not using any promoters at the moment!' 74 | # promoters = np.zeros((enhancers.shape[0], 4, 0)) 75 | 76 | print 'Loading promoter data from ' + promoter_f_name + '...' 77 | promoters = load_file(promoter_f_name) 78 | 79 | return np.concatenate((enhancers, promoters), 2) 80 | 81 | # Reads the specified files and builds num_samples X 4 X seg_length arrays for 82 | # both the positive and negative data sets; returns concatenations of positive 83 | # and negative data sets, as well as the labels 84 | def load_pn_data(positive_f_name, negative_f_name): 85 | 86 | print 'Loading positive data from ' + positive_f_name + '...' 87 | positive = load_file(positive_f_name) 88 | positive_labels = np.ones(positive.shape[0]) 89 | 90 | print 'Loading negative data from ' + negative_f_name + '...' 91 | negative = load_file(negative_f_name) 92 | negative_labels = np.zeros(negative.shape[0]) 93 | 94 | samples = np.concatenate((positive, negative), 0) 95 | labels = np.concatenate((positive_labels, negative_labels), 0) 96 | return samples, labels 97 | 98 | # Reads the specified file and returns a 3D 99 | # num_samples X 4 X segment_length array 100 | def load_file(f_name): 101 | 102 | num_pairs = sum(1 for line in open(f_name))/4; # number of sample pairs 103 | 104 | # Declare output var but can't allocate space till we know segment_length 105 | Xs = None 106 | 107 | with open(f_name, 'r') as f: 108 | 109 | line_num = 0 # number of lines (i.e., samples) read so far 110 | for line in f.read().splitlines(): 111 | 112 | if line_num == 0: 113 | # allocate space for output 114 | Xs = np.zeros((num_pairs, 4, len(line))) 115 | 116 | sample_type = line_num % 4; # 0, 1, 2, or 4, denoting A, T, C, or G 117 | sample_num = line_num / 4; 118 | 119 | Xs[sample_num, sample_type, :] = [int(x) for x in line]; 120 | 121 | line_num += 1 122 | 123 | return Xs 124 | -------------------------------------------------------------------------------- /pairwise/data_processing/txt_to_hdf5_ep_split.py: -------------------------------------------------------------------------------- 1 | # Basic python and data processing imports 2 | import numpy as np 3 | import h5py 4 | from sklearn.utils import shuffle 5 | import load_data_pairs as ld # my own scripts for loading data 6 | 7 | # input data paths 8 | cell_line = '4lines' 9 | data_prefix = '/home/sss1/Desktop/projects/DeepInteractions/data/uniform_len/' + cell_line + '/' + cell_line 10 | positive_enhancers = data_prefix + '_pos_1.txt' 11 | positive_promoters = data_prefix + '_pos_2.txt' 12 | positive_enhancers_aug = data_prefix + '_pos_1_aug.txt' 13 | positive_promoters_aug = data_prefix + '_pos_2_aug.txt' 14 | negative_enhancers = data_prefix + '_neg_1.txt' 15 | negative_promoters = data_prefix + '_neg_2.txt' 16 | 17 | # Save state of random number generator so we can jointly shuffle data 18 | rng_state = np.random.get_state() 19 | 20 | # output data path 21 | out_path = data_prefix + '_ep_split.h5' 22 | 23 | with h5py.File(out_path + 'ori_enh', 'w') as hf: 24 | 25 | print 'Loading original enhancers...' 26 | X_enhancers_train, y_train = ld.load_pn_data(positive_enhancers, negative_enhancers) 27 | X_enhancers_train = np.transpose(X_enhancers_train, axes=(0,2,1)) 28 | print 'Shuffling original enhancers...' 29 | np.random.set_state(rng_state) 30 | np.random.shuffle(X_enhancers_train) 31 | np.random.set_state(rng_state) 32 | np.random.shuffle(y_train) 33 | print 'Saving original enhancers...' 34 | hf.create_dataset('X_enhancers_train', data = X_enhancers_train) 35 | hf.create_dataset('y_train', data = y_train) 36 | 37 | with h5py.File(out_path + 'ori_pro', 'w') as hf: 38 | 39 | print 'Loading original promoters...' 40 | X_promoters_train, _ = ld.load_pn_data(positive_promoters, negative_promoters) 41 | X_promoters_train = np.transpose(X_promoters_train, axes=(0,2,1)) 42 | print 'Shuffling original promoters...' 43 | np.random.set_state(rng_state) 44 | np.random.shuffle(X_promoters_train) 45 | print 'Saving original promoters...' 46 | hf.create_dataset('X_promoters_train', data = X_promoters_train) 47 | 48 | with h5py.File(out_path + 'aug_enh', 'w') as hf: 49 | 50 | print 'Loading augmented enhancers...' 51 | X_enhancers_train_aug, y_train_aug = ld.load_pn_data(positive_enhancers_aug, negative_enhancers) 52 | X_enhancers_train_aug = np.transpose(X_enhancers_train_aug, axes=(0,2,1)) 53 | print 'Shuffling augmented enhancers...' 54 | np.random.set_state(rng_state) 55 | np.random.shuffle(X_enhancers_train_aug) 56 | np.random.set_state(rng_state) 57 | np.random.shuffle(y_train_aug) 58 | print 'Saving augmented enhancers...' 59 | hf.create_dataset('X_enhancers_train_aug', data = X_enhancers_train_aug) 60 | hf.create_dataset('y_train_aug', data = y_train_aug) 61 | 62 | with h5py.File(out_path + 'aug_pro', 'w') as hf: 63 | 64 | print 'Loading augmented promoters...' 65 | X_promoters_train_aug, _ = ld.load_pn_data(positive_promoters_aug, negative_promoters) 66 | X_promoters_train_aug = np.transpose(X_promoters_train_aug, axes=(0,2,1)) 67 | print 'Shuffling augmented promoters...' 68 | np.random.set_state(rng_state) 69 | np.random.shuffle(X_promoters_train_aug) 70 | print 'Saving augmented promoters...' 71 | hf.create_dataset('X_promoters_train_aug', data = X_promoters_train_aug) 72 | -------------------------------------------------------------------------------- /pairwise/data_processing/write_predictions_to_CSV.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | import h5py 3 | from sklearn.metrics import precision_recall_curve, roc_curve, auc, average_precision_score 4 | # import matplotlib.pyplot as plt 5 | 6 | from keras.optimizers import Adam # not used but needed to compile model 7 | 8 | test_cell_lines = ['GM12878', 'HeLa-S3', 'HUVEC', 'IMR90', 'K562', 'NHEK'] 9 | for test_cell_line in test_cell_lines: 10 | 11 | data_path = '/home/sss1/Desktop/projects/DeepInteractions/data/uniform_len/' + test_cell_line + '/' + test_cell_line + '_ep_split.h5' 12 | predictions_path = '/home/sss1/Desktop/projects/DeepInteractions/data/uniform_len/' + test_cell_line + '/train_small' + test_cell_line + '_test_' + test_cell_line + '_predictions.h5' 13 | out_path = '/home/sss1/Desktop/projects/DeepInteractions/pairwise/tmp_data/' + test_cell_line + '.csv' 14 | 15 | # print 'Loading labels from ' + data_path + '...' 16 | with h5py.File(data_path, 'r') as hf: 17 | y = np.array(hf.get('y_train')) 18 | # print 'Loading predictions from ' + predictions_path + '...' 19 | with h5py.File(predictions_path, 'r') as hf: 20 | y_score = np.squeeze(np.array(hf.get('y_score'))) 21 | 22 | # print 'y: ' + str(y) 23 | # print 'y.shape(): ' + str(np.shape(y)) 24 | # # print 'mean(y) :' + str(np.mean(y)) 25 | # print 'y_score: ' + str(y_score) 26 | # print 'y_score.shape(): ' + str(np.shape(y_score)) 27 | # # print 'mean(y_score) :' + str(np.mean(y_score)) 28 | # print 'corrcoef(y, y_score) :' + str(np.corrcoef(y, y_score)) 29 | 30 | # precision, recall, thresholds = precision_recall_curve(y, y_score) 31 | print '\nCell line: ' + test_cell_line 32 | fpr, tpr, _ = roc_curve(y, y_score) 33 | print 'AUPR: ' + str(average_precision_score(y, y_score)) 34 | print 'AUROC: ' + str(auc(fpr, tpr)) 35 | # 36 | # plt.clf() 37 | # # plt.plot(fpr, tpr) 38 | # plt.plot(recall, precision) 39 | # plt.xlabel('Recall') 40 | # plt.ylabel('Precision') 41 | # plt.ylim([0.0, 1.05]) 42 | # plt.xlim([0.0, 1.0]) 43 | # plt.show() 44 | 45 | output = np.transpose(np.vstack((y, y_score))) 46 | print 'Combined shape:' + str(np.shape(output)) 47 | 48 | np.savetxt(out_path, output, delimiter = ",") 49 | -------------------------------------------------------------------------------- /pairwise/data_processing/write_sequences_to_fasta.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | import load_data_pairs as ld 3 | import h5py 4 | 5 | root = '/home/sss1/Desktop/projects/DeepInteractions/data/uniform_len/original/' 6 | data_path = root + 'all_data.h5' 7 | cell_lines = ['GM12878', 'HeLa-S3', 'HUVEC', 'IMR90', 'K562', 'NHEK'] 8 | 9 | # Map: 10 | # [1,0,0,0] -> A 11 | # [0,1,0,0] -> T 12 | # [0,0,1,0] -> C 13 | # [0,0,0,1] -> G 14 | def one_hot_to_letter(base): 15 | if not (np.sum(base) == 1 and np.min(base) == 0): 16 | return 'N' 17 | if base[0] == 1: 18 | return 'A' 19 | if base[1] == 1: 20 | return 'T' 21 | if base[2] == 1: 22 | return 'C' 23 | if base[3] == 1: 24 | return 'G' 25 | return 'N' 26 | 27 | # data: (num_sequences X sequence_length X 4) 3-tensor of num_sequences 28 | # one-hot encoded nucleotide sequences of equal-length sequence_length 29 | # name: string label for the data set (e.g., 'K562_enhancers') 30 | # path: string file path to which to print the data 31 | def format_file(data, name): 32 | print 'Formatting ' + name + ' data ...' 33 | file_to_print = '' 34 | sequence_idx = 0 35 | for sequence in data: 36 | sequence_to_print = '' 37 | for base in sequence: 38 | sequence_to_print += str(one_hot_to_letter(base)) 39 | 40 | file_to_print += '>' + str(sequence_idx) + '\n' 41 | file_to_print += sequence_to_print + '\n' 42 | sequence_idx += 1 43 | 44 | return file_to_print 45 | 46 | 47 | with h5py.File(data_path, 'r') as hf: 48 | 49 | for cell_line in cell_lines: 50 | print 'Loading ' + cell_line + ' data from ' + data_path 51 | 52 | # Print enhancer data 53 | X_enhancers = np.array(hf.get(cell_line + '_X_enhancers')).transpose((0, 2, 1)) 54 | name = cell_line + '_enhancers' 55 | out_path = root + 'asFASTA/' + name + '.fasta' 56 | file_contents = format_file(X_enhancers, name) 57 | print 'Writing ' + name + ' data to ' + out_path + ' ...' 58 | f = open(out_path, 'w') 59 | f.write(file_contents) 60 | f.close() 61 | 62 | # Print promoter data 63 | X_promoters = np.array(hf.get(cell_line + '_X_promoters')).transpose((0, 2, 1)) 64 | name = cell_line + '_promoters' 65 | out_path = root + 'asFASTA/' + name + '.fasta' 66 | file_contents = format_file(X_promoters, name) 67 | print 'Writing ' + name + ' data to ' + out_path + ' ...' 68 | f = open(out_path, 'w') 69 | f.write(file_contents) 70 | f.close() 71 | -------------------------------------------------------------------------------- /pairwise/fimo_all_GM12878_motifs.sh: -------------------------------------------------------------------------------- 1 | fimo -oc NHEK_enhancers_small_retest ~/Desktop/motif_databases/motif_TargetFinder/GM12878_targetfinderMatchP.meme ../../data/uniform_len/original/asFASTA/NHEK_enhancers.fasta & 2 | fimo -oc IMR90_enhancers_small_retest ~/Desktop/motif_databases/motif_TargetFinder/GM12878_targetfinderMatchP.meme ../../data/uniform_len/original/asFASTA/IMR90_enhancers.fasta & 3 | fimo -oc HUVEC_enhancers_small_retest ~/Desktop/motif_databases/motif_TargetFinder/GM12878_targetfinderMatchP.meme ../../data/uniform_len/original/asFASTA/HUVEC_enhancers.fasta & 4 | fimo -oc GM12878_enhancers_small_retest ~/Desktop/motif_databases/motif_TargetFinder/GM12878_targetfinderMatchP.meme ../../data/uniform_len/original/asFASTA/GM12878_enhancers.fasta 5 | 6 | fimo -oc HeLa-S3_enhancers_small_retest ~/Desktop/motif_databases/motif_TargetFinder/GM12878_targetfinderMatchP.meme ../../data/uniform_len/original/asFASTA/HeLa-S3_enhancers.fasta & 7 | fimo -oc NHEK_promoters_small_retest ~/Desktop/motif_databases/motif_TargetFinder/GM12878_targetfinderMatchP.meme ../../data/uniform_len/original/asFASTA/NHEK_promoters.fasta & 8 | fimo -oc IMR90_promoters_small_retest ~/Desktop/motif_databases/motif_TargetFinder/GM12878_targetfinderMatchP.meme ../../data/uniform_len/original/asFASTA/IMR90_promoters.fasta & 9 | fimo -oc K562_enhancers_small_retest ~/Desktop/motif_databases/motif_TargetFinder/GM12878_targetfinderMatchP.meme ../../data/uniform_len/original/asFASTA/K562_enhancers.fasta 10 | 11 | fimo -oc HUVEC_promoters_small_retest ~/Desktop/motif_databases/motif_TargetFinder/GM12878_targetfinderMatchP.meme ../../data/uniform_len/original/asFASTA/HUVEC_promoters.fasta & 12 | fimo -oc GM12878_promoters_small_retest ~/Desktop/motif_databases/motif_TargetFinder/GM12878_targetfinderMatchP.meme ../../data/uniform_len/original/asFASTA/GM12878_promoters.fasta & 13 | fimo -oc K562_promoters_small_retest ~/Desktop/motif_databases/motif_TargetFinder/GM12878_targetfinderMatchP.meme ../../data/uniform_len/original/asFASTA/K562_promoters.fasta & 14 | fimo -oc HeLa-S3_promoters_small_retest ~/Desktop/motif_databases/motif_TargetFinder/GM12878_targetfinderMatchP.meme ../../data/uniform_len/original/asFASTA/HeLa-S3_promoters.fasta 15 | -------------------------------------------------------------------------------- /pairwise/frozen_model/build_model.py: -------------------------------------------------------------------------------- 1 | import util 2 | # Keras imports 3 | from keras.layers import Input, Convolution1D, MaxPooling1D, Merge, Dropout, Flatten, Dense, BatchNormalization, LSTM, Activation, Bidirectional 4 | from keras.optimizers import RMSprop, Adam 5 | from keras.callbacks import ModelCheckpoint, EarlyStopping, Callback 6 | from keras.models import Sequential 7 | # from seq2seq.layers.bidirectional import Bidirectional 8 | from keras.regularizers import l2, activity_l2 9 | 10 | # model parameters 11 | enhancer_length = 3000 # TODO: get this from input 12 | promoter_length = 2000 # TODO: get this from input 13 | n_kernels = 1024 # Number of kernels; used to be 1024 14 | filter_length = 40 # Length of each kernel 15 | LSTM_out_dim = 100 # Output direction of ONE DIRECTION of LSTM; used to be 512 16 | 17 | 18 | # Convolutional/maxpooling layers to extract prominent motifs 19 | # Separate identically initialized convolutional layers are trained for 20 | # enhancers and promoters 21 | # Define enhancer layers 22 | enhancer_conv_layer = Convolution1D(input_dim = 4, 23 | input_length = enhancer_length, 24 | nb_filter = n_kernels, 25 | filter_length = filter_length, 26 | border_mode = "valid", 27 | subsample_length = 1, 28 | W_regularizer = l2(1e-6)) 29 | enhancer_max_pool_layer = MaxPooling1D(pool_length = filter_length/2, stride = filter_length/2) 30 | 31 | # Build enhancer branch 32 | enhancer_branch = Sequential() 33 | enhancer_branch.add(enhancer_conv_layer) 34 | # enhancer_branch.add(Activation("relu")) 35 | enhancer_branch.add(enhancer_max_pool_layer) 36 | 37 | # Define promoter layers branch: 38 | promoter_conv_layer = Convolution1D(input_dim = 4, 39 | input_length = promoter_length, 40 | nb_filter = n_kernels, 41 | filter_length = filter_length, 42 | border_mode = "valid", 43 | subsample_length = 1, 44 | W_regularizer = l2(1e-6)) 45 | promoter_max_pool_layer = MaxPooling1D(pool_length = filter_length/2, stride = filter_length/2) 46 | 47 | # Build promoter branch 48 | promoter_branch = Sequential() 49 | promoter_branch.add(promoter_conv_layer) 50 | # promoter_branch.add(Activation("relu")) 51 | promoter_branch.add(promoter_max_pool_layer) 52 | 53 | # Define main model layers 54 | # Concatenate outputs of enhancer and promoter convolutional layers 55 | merge_layer = Merge([enhancer_branch, promoter_branch], 56 | mode = 'concat', 57 | concat_axis = 1) 58 | 59 | 60 | # Bidirectional LSTM to extract combinations of motifs 61 | biLSTM_layer = Bidirectional(LSTM(input_dim = n_kernels, 62 | output_dim = LSTM_out_dim, 63 | return_sequences = True)) 64 | 65 | # Dense layer to allow nonlinearities 66 | dense_layer = Dense(output_dim = 1000, 67 | init = "glorot_uniform", 68 | W_regularizer = l2(1e-6)) 69 | 70 | # Logistic regression layer to make final binary prediction 71 | LR_classifier_layer = Dense(output_dim = 1) 72 | 73 | 74 | def build_model(use_JASPAR = True): 75 | 76 | # A single downstream model merges the enhancer and promoter branches 77 | # Build main (merged) branch 78 | # Using batch normalization seems to inhibit retraining, probably because the 79 | # point of retraining is to learn (external) covariate shift 80 | model = Sequential() 81 | model.add(merge_layer) 82 | model.add(BatchNormalization()) 83 | model.add(Dropout(0.25)) 84 | model.add(biLSTM_layer) 85 | model.add(BatchNormalization()) 86 | model.add(Dropout(0.5)) 87 | model.add(Flatten()) 88 | model.add(dense_layer) 89 | model.add(BatchNormalization()) 90 | model.add(Activation("relu")) 91 | model.add(Dropout(0.5)) 92 | model.add(LR_classifier_layer) 93 | model.add(BatchNormalization()) 94 | model.add(Activation("sigmoid")) 95 | 96 | # Read in and initialize convolutional layers with motifs from JASPAR 97 | if use_JASPAR: 98 | util.initialize_with_JASPAR(enhancer_conv_layer, promoter_conv_layer) 99 | 100 | return model 101 | 102 | def build_frozen_model(): 103 | 104 | # Freeze all by the dense layers of the network 105 | enhancer_conv_layer.trainable = False 106 | enhancer_max_pool_layer.trainable = False 107 | promoter_conv_layer.trainable = False 108 | promoter_max_pool_layer.trainable = False 109 | biLSTM_layer.trainable = False 110 | 111 | # TODO: Figure out how to remove layers after loading weights 112 | 113 | model = Sequential() 114 | model.add(merge_layer) 115 | model.add(BatchNormalization()) 116 | model.add(Dropout(0.25)) 117 | model.add(biLSTM_layer) 118 | model.add(BatchNormalization()) 119 | model.add(Dropout(0.5)) 120 | model.add(Flatten()) 121 | model.add(dense_layer) 122 | model.add(BatchNormalization()) 123 | model.add(Activation("relu")) 124 | model.add(Dropout(0.5)) 125 | model.add(LR_classifier_layer) 126 | model.add(BatchNormalization()) 127 | model.add(Activation("sigmoid")) 128 | 129 | return model 130 | -------------------------------------------------------------------------------- /pairwise/frozen_model/retraining_model3.py: -------------------------------------------------------------------------------- 1 | # Basic python and data processing imports 2 | import numpy as np 3 | np.set_printoptions(suppress=True) # Suppress scientific notation when printing small 4 | import h5py 5 | import scipy.io 6 | from datetime import datetime 7 | import load_data_pairs as ld # my own scripts for loading data 8 | import build_model 9 | 10 | # Save plot of errors, but don't display it live 11 | # import matplotlib 12 | # matplotlib.use('Agg') 13 | import matplotlib.pyplot as plt 14 | import util 15 | # np.random.seed(1337) # for reproducibility 16 | 17 | # Keras imports 18 | from keras.layers import Input, Convolution1D, MaxPooling1D, Merge, Dropout, Flatten, Dense, BatchNormalization, LSTM, Activation 19 | from keras.optimizers import RMSprop, Adam 20 | from keras.callbacks import ModelCheckpoint, EarlyStopping, Callback# , ReduceLROnPlateau 21 | from keras.models import Sequential 22 | # from keras.utils.visualize_util import plot 23 | from seq2seq.layers.bidirectional import Bidirectional 24 | 25 | # TODO: Refactor model into build_model.py 26 | 27 | # training parameters 28 | cell_line = 'union' 29 | num_epochs = 32 30 | num_epochs_frozen = 360 31 | batch_size = 50 32 | training_frac = 0.9 # use 90% of data for training, 10% for testing/validation 33 | t = datetime.now().strftime('%Y-%m-%d-%H:%M:%S') 34 | checkpoint_path = "/home/sss1/Desktop/projects/DeepInteractions/weights/" + cell_line + "-JASPAR-" + t + ".hdf5" 35 | opt = Adam(lr = 1e-5) # opt = RMSprop(lr = 1e-6) 36 | 37 | # # Load data and split into training and validation sets 38 | data_path = 39 | '/home/sss1/Desktop/projects/DeepInteractions/data/uniform_len/original/all_data.h5' 40 | print 'Loading data from ' + data_path 41 | # TODO: Resample 10 times and do cross-validation 42 | X_enhancer_train, X_promoter_train, y_train = ld.load_hdf5_ep_split_aug(data_path) 43 | X_enhancer_train, X_promoter_train, y_train, X_enhancer_val, X_promoter_val, y_val = util.split_train_and_val_data(X_enhancer_train, X_promoter_train, y_train, training_frac) 44 | 45 | enhancer_length = X_enhancer_train.shape[1] 46 | promoter_length = X_promoter_train.shape[1] 47 | 48 | print 'Building model...' 49 | model = build_model.build_model() 50 | 51 | # model.load_weights('/home/sss1/Desktop/projects/DeepInteractions/weights/myDanQ-JASPAR_bestmodel-2016-10-08-01:02:54.hdf5') 52 | 53 | print 'Compiling model...' 54 | model.compile(loss = 'binary_crossentropy', 55 | optimizer = opt, 56 | metrics = ["accuracy"]) 57 | 58 | # Print a summary of the model 59 | model.summary() 60 | 61 | # Define custom callback that prints/plots performance at end of each epoch 62 | class ConfusionMatrix(Callback): 63 | def on_train_begin(self, logs = {}): 64 | self.epoch = 0 65 | self.precisions = [] 66 | self.recalls = [] 67 | self.f1_scores = [] 68 | self.losses = [] 69 | self.training_losses = [] 70 | self.training_accs = [] 71 | self.accs = [] 72 | plt.ion() 73 | 74 | # def on_batch_end(self, batch, logs = {}): 75 | # self.training_losses.append(logs.get('loss')) 76 | # self.training_accs.append(logs.get('acc')) 77 | 78 | def on_epoch_end(self, batch, logs = {}): 79 | self.training_losses.append(logs.get('loss')) 80 | self.training_accs.append(logs.get('acc')) 81 | self.epoch += 1 82 | val_predict = model.predict_classes([X_enhancer_val, X_promoter_val], batch_size = batch_size, verbose = 0) 83 | util.print_live(self, y_val, val_predict, logs) 84 | if self.epoch > 1: # need at least two time points to plot 85 | util.plot_live(self) 86 | 87 | print '\nData sizes: ' 88 | print '[X_enhancers_train, X_promoters_train]: [' + str(np.shape(X_enhancer_train)) + ', ' + str(np.shape(X_promoter_train)) + ']' 89 | print 'y_train: ' + str(np.shape(y_train)) 90 | print 'y_train.mean(): ' + str(y_train.mean()) 91 | 92 | # Instantiate callbacks 93 | confusionMatrix = ConfusionMatrix() 94 | checkpointer = ModelCheckpoint(filepath=checkpoint_path, 95 | verbose=1, 96 | save_best_only=True) 97 | 98 | # # Reduce learning rate by 1/5 if val_loss is stagnant for 10 epochs 99 | # reduce_lr = ReduceLROnPlateau(monitor='val_loss', 100 | # factor=0.2, 101 | # patience=10) 102 | # earlystopper = EarlyStopping(monitor='val_loss', patience=20, verbose=1) 103 | print 'Running fully trainable model for exactly ' + str(num_epochs) + ' epochs...' 104 | model.fit([X_enhancer_train, X_promoter_train], 105 | [y_train], 106 | validation_data = ([X_enhancer_val, X_promoter_val], y_val), 107 | batch_size = batch_size, 108 | nb_epoch = num_epochs, 109 | shuffle = True, 110 | callbacks=[confusionMatrix, checkpointer]# , reduce_lr] 111 | ) 112 | 113 | plotName = cell_line + '_' + t + '.png' 114 | plt.savefig(plotName) 115 | print 'Saved loss plot to ' + plotName 116 | 117 | ### BEGIN CODE FOR RETRAINING MODEL ON IMBALANCED DATA ### 118 | print '\n\n\n\nBEGINNING RETRAINING PHASE.\n\n\n\n' 119 | 120 | # Rebuild the model. The purpose of this is mostly to reset/remove the batch 121 | # normalization layers (there's probably a better way)... 122 | print 'Building frozen model...' 123 | model = build_model.build_frozen_model() 124 | 125 | print 'Compiling retraining model...' 126 | model.compile(loss = 'binary_crossentropy', 127 | optimizer = opt, 128 | metrics=["accuracy"]) 129 | 130 | # subsample balanced training data to create imbalanced training data 131 | X_enhancer_train, X_promoter_train, y_train = ld.load_hdf5_ep_split(data_path) 132 | X_enhancer_train, X_promoter_train, y_train, X_enhancer_val, X_promoter_val, y_val = util.split_train_and_val_data(X_enhancer_train, X_promoter_train, y_train, training_frac) 133 | 134 | # fraction of samples in each class 135 | pos_frac = y_train.mean() 136 | neg_weight = (1/(1 - pos_frac))**(1/2) 137 | pos_weight = (1/pos_frac)**(1/2) 138 | 139 | print 'Positive weight: ' + str(pos_weight) + ' Negative weight: ' + str(neg_weight) 140 | 141 | # Instantiate callbacks for frozen training 142 | confusionMatrixFrozen = ConfusionMatrix() 143 | # # Reduce learning rate by 1/5 if val_loss is stagnant for 10 epochs 144 | # reduce_lr = ReduceLROnPlateau(monitor='val_loss', 145 | # factor=0.2, 146 | # patience=10) 147 | print 'Running partly frozen model for exactly ' + str(num_epochs_frozen) + ' epochs...' 148 | model.fit([X_enhancer_train, X_promoter_train], 149 | [y_train], 150 | validation_data = ([X_enhancer_val, X_promoter_val], y_val), 151 | batch_size = batch_size, 152 | nb_epoch = num_epochs_frozen, 153 | shuffle = True, 154 | callbacks=[confusionmatrixfrozen],# , reduce_lr], 155 | class_weight = {0 : neg_weight, 1 : pos_weight} # increase weight of positive samples, to counter class imbalance 156 | ) 157 | 158 | plotName = cell_line + '_frozen_' + t + '.png' 159 | plt.savefig(plotName) 160 | print 'Saved loss plot to ' + plotName 161 | -------------------------------------------------------------------------------- /pairwise/load_data_pairs.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | import sys 3 | import h5py 4 | 5 | def load_hdf5(path): 6 | with h5py.File(path,'r') as hf: 7 | X_train = np.array(hf.get('X_train')) 8 | y_train = np.array(hf.get('y_train')) 9 | return X_train, y_train 10 | 11 | def load_hdf5_ep_split(path): 12 | with h5py.File(path,'r') as hf: 13 | X_enhancers_train = np.array(hf.get('X_enhancers_train')) 14 | X_promoters_train = np.array(hf.get('X_promoters_train')) 15 | y_train = np.array(hf.get('y_train')) 16 | return X_enhancers_train, X_promoters_train, y_train 17 | 18 | def load_hdf5_ep_split_aug(path): 19 | with h5py.File(path,'r') as hf: 20 | X_enhancers_train = np.array(hf.get('X_enhancers_train_aug')) 21 | X_promoters_train = np.array(hf.get('X_promoters_train_aug')) 22 | y_train = np.array(hf.get('y_train_aug')) 23 | return X_enhancers_train, X_promoters_train, y_train 24 | 25 | # Reads the specified files and builds num_samples X 4 X seg_length arrays for 26 | # both positive and negative versions of the enhancers and promoter data sets 27 | # Outputs: 28 | # 1) concatenations of enhancer and promoter segments, for all positive and 29 | # negative samples 30 | # 2) labels (0 for negative, 1 for positive) for corresponding samples 31 | def load_full_data(pos_enhancer_f_name, pos_promoter_f_name, neg_enhancer_f_name, neg_promoter_f_name): 32 | 33 | print 'Loading positive samples...' 34 | positive = load_ep_pairs(pos_enhancer_f_name, pos_promoter_f_name) 35 | positive_labels = np.ones(positive.shape[0]) 36 | 37 | print 'Loading negative samples...' 38 | negative = load_ep_pairs(neg_enhancer_f_name, neg_promoter_f_name) 39 | negative_labels = np.zeros(negative.shape[0]) 40 | 41 | samples = np.concatenate((positive, negative), 0) 42 | labels = np.concatenate((positive_labels, negative_labels), 0) 43 | return samples, labels 44 | 45 | # TODO!! 46 | def load_imbalanced_data(pos_neg_ratio, pos_enhancer_f_name, pos_promoter_f_name, neg_enhancer_f_name, neg_promoter_f_name): 47 | 48 | print 'Loading positive samples...' 49 | positive = load_ep_pairs(pos_enhancer_f_name, pos_promoter_f_name) 50 | n_pos = positive.shape[0] # original number of positive samples 51 | n_pos_sub = round(n_pos * pos_neg_ratio) # number of subsampled positive samples 52 | 53 | positive_subsampled = positive[np.random.choice(n_pos, n_pos_sub, replace=False),:, :] 54 | positive_labels = np.ones(n_pos_sub) 55 | 56 | print 'Loading negative samples...' 57 | negative = load_ep_pairs(neg_enhancer_f_name, neg_promoter_f_name) 58 | negative_labels = np.zeros(negative.shape[0]) 59 | 60 | samples = np.concatenate((positive_subsampled, negative), 0) 61 | labels = np.concatenate((positive_labels, negative_labels), 0) 62 | return samples, labels 63 | 64 | # Reads the specified files and builds num_samples X 4 X seg_length arrays for 65 | # both the enhancers and promoter data sets; returns concatenations of enhancer 66 | # and promoter segments 67 | def load_ep_pairs(enhancer_f_name, promoter_f_name): 68 | 69 | print 'Loading enhancer data from ' + enhancer_f_name + '...' 70 | enhancers = load_file(enhancer_f_name) 71 | 72 | # # Code to just load enhancers 73 | # print 'Not using any promoters at the moment!' 74 | # promoters = np.zeros((enhancers.shape[0], 4, 0)) 75 | 76 | print 'Loading promoter data from ' + promoter_f_name + '...' 77 | promoters = load_file(promoter_f_name) 78 | 79 | return np.concatenate((enhancers, promoters), 2) 80 | 81 | # Reads the specified files and builds num_samples X 4 X seg_length arrays for 82 | # both the positive and negative data sets; returns concatenations of positive 83 | # and negative data sets, as well as the labels 84 | def load_pn_data(positive_f_name, negative_f_name): 85 | 86 | print 'Loading positive data from ' + positive_f_name + '...' 87 | positive = load_file(positive_f_name) 88 | positive_labels = np.ones(positive.shape[0]) 89 | 90 | print 'Loading negative data from ' + negative_f_name + '...' 91 | negative = load_file(negative_f_name) 92 | negative_labels = np.zeros(negative.shape[0]) 93 | 94 | samples = np.concatenate((positive, negative), 0) 95 | labels = np.concatenate((positive_labels, negative_labels), 0) 96 | return samples, labels 97 | 98 | # Reads the specified file and returns a 3D 99 | # num_samples X 4 X segment_length array 100 | def load_file(f_name): 101 | 102 | num_pairs = sum(1 for line in open(f_name))/4; # number of sample pairs 103 | 104 | # Declare output var but can't allocate space till we know segment_length 105 | Xs = None 106 | 107 | with open(f_name, 'r') as f: 108 | 109 | line_num = 0 # number of lines (i.e., samples) read so far 110 | for line in f.read().splitlines(): 111 | 112 | if line_num == 0: 113 | # allocate space for output 114 | Xs = np.zeros((num_pairs, 4, len(line))) 115 | 116 | sample_type = line_num % 4; # 0, 1, 2, or 4, denoting A, T, C, or G 117 | sample_num = line_num / 4; 118 | 119 | Xs[sample_num, sample_type, :] = [int(x) for x in line]; 120 | 121 | line_num += 1 122 | 123 | return Xs 124 | -------------------------------------------------------------------------------- /pairwise/predict.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | import h5py 3 | import matplotlib.pyplot as plt 4 | import util 5 | 6 | import build_small_model as bm 7 | import load_data_pairs as ld # my own scripts for loading data 8 | 9 | from keras.optimizers import Adam # not used but needed to compile model 10 | 11 | def predict(train_cell_line, test_cell_line): 12 | 13 | data_path = '/home/sss1/Desktop/projects/DeepInteractions/data/uniform_len/' + test_cell_line + '/' + test_cell_line + '_ep_split.h5' 14 | predictions_path = '/home/sss1/Desktop/projects/DeepInteractions/data/uniform_len/' + test_cell_line + '/train_small' + train_cell_line + '_test_' + test_cell_line + '_predictions.h5' 15 | 16 | print 'Loading test data...' 17 | X_enhancer, X_promoter, y = ld.load_hdf5_ep_split(data_path) 18 | 19 | print 'Building model...' 20 | model = bm.build_model(use_JASPAR = False) 21 | 22 | print 'Compiling model...' 23 | opt = Adam(lr = 1e-5) 24 | model.compile(loss = 'binary_crossentropy', 25 | optimizer = opt, 26 | metrics = ["accuracy"]) 27 | 28 | print 'Loading ' + train_cell_line + ' model weights...' 29 | model.load_weights('/home/sss1/Desktop/projects/DeepInteractions/weights/best/small_model_balanced' + train_cell_line + '-noJASPAR.hdf5') 30 | 31 | print 'Running predictions...' 32 | y_score = model.predict([X_enhancer, X_promoter], batch_size = 50, verbose = 1) 33 | 34 | print 'Saving predictions...' 35 | with h5py.File(predictions_path, 'w') as hf: 36 | hf.create_dataset('y_score', data = y_score) 37 | print 'Saved predictions to ' + predictions_path 38 | 39 | 40 | def plot_PR_and_ROC_curves(train_cell_line, test_cell_line): 41 | data_path = '/home/sss1/Desktop/projects/DeepInteractions/data/uniform_len/' + test_cell_line + '/' + test_cell_line + '_ep_split.h5' 42 | predictions_path = '/home/sss1/Desktop/projects/DeepInteractions/data/uniform_len/' + test_cell_line + '/train_small' + train_cell_line + '_test_' + test_cell_line + '_predictions.h5' 43 | print 'True label path: ' + data_path 44 | print 'Prediction score path: ' + predictions_path 45 | 46 | # # print 'Loading labels from ' + data_path + '...' 47 | # with h5py.File(data_path, 'r') as hf: 48 | # y = np.array(hf.get('y_train')) 49 | # # print 'Loading predictions from ' + predictions_path + '...' 50 | # with h5py.File(predictions_path, 'r') as hf: 51 | # y_score = np.array(hf.get('y_score')) 52 | 53 | # # ap = util.plot_PR_curve(y, y_score) 54 | # # print '(' + train_cell_line + ', ' + test_cell_line + ') AUPR: ' + str(round(ap, 2)) 55 | # roc_auc = util.plot_ROC_curve(y, y_score) 56 | # print '(' + train_cell_line + ', ' + test_cell_line + ') AUROC: ' + str(round(roc_auc, 2)) 57 | 58 | train_cell_lines = ['4lines']# ['GM12878', 'HeLa-S3', 'HUVEC', 'IMR90', 'K562', 'NHEK'] 59 | test_cell_lines = ['GM12878', 'HeLa-S3', 'HUVEC', 'IMR90', 'K562', 'NHEK'] 60 | for train_cell_line in train_cell_lines: 61 | for test_cell_line in test_cell_lines: 62 | plot_PR_and_ROC_curves(train_cell_line, test_cell_line) 63 | # print '\nPredicting ' + test_cell_line + ' after training on ' + train_cell_line 64 | # predict(train_cell_line, test_cell_line) 65 | print '' 66 | -------------------------------------------------------------------------------- /pairwise/read_FIMO_results.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | import csv 3 | from sklearn.metrics import average_precision_score 4 | from keras.optimizers import Adam # needed to compile prediction model 5 | import h5py 6 | import load_data_pairs as ld # my own scripts for loading data 7 | import build_small_model as bm 8 | import util 9 | 10 | fimo_root = '/home/sss1/Desktop/projects/DeepInteractions/pairwise/FIMO/' 11 | data_path = '/home/sss1/Desktop/projects/DeepInteractions/data/uniform_len/original/all_data.h5' 12 | 13 | cell_lines = ['GM12878', 'HeLa-S3', 'HUVEC', 'IMR90', 'K562', 'NHEK'] 14 | data_types = ['enhancers', 'promoters'] 15 | 16 | # num_repeats = 5 # number of i.i.d. trials to run; too slow to do :( 17 | random_window_length = 20 # number of bp to randomize at each feature occurence 18 | 19 | # Randomize each appearance of the pattern pattern in the data 20 | def randomize_window(sequence): 21 | for base_idx in range(np.shape(sequence)[0]): 22 | sequence[base_idx] = np.zeros(4) 23 | sequence[base_idx, np.random.randint(0, 4)] = 1 24 | 25 | # Returns a deep copy of the data, with motif occurrences randomized out. 26 | # A deep copy is made because this is much faster than reloading the data for 27 | # every motif. 28 | # data: (num_sequences X sequence_length X 4) 3-tensor of num_sequences 29 | # one-hot encoded nucleotide sequences of equal-length sequence_length 30 | # motifs_idxs: list of (sample_idx, start_idx, stop_idx) triples 31 | def replace_motifs_in_data(data, motif_idxs): 32 | data_copy = np.copy(data) 33 | for (sample_idx, motif_start, motif_stop) in idxs: 34 | mid = (motif_start + motif_stop)/2 35 | start = max(0, mid - (random_window_length/2)) 36 | stop = min(np.shape(data)[1], start + random_window_length) 37 | randomize_window(data_copy[sample_idx, start:stop, :]) 38 | return data_copy 39 | 40 | for cell_line in cell_lines: 41 | for data_type in data_types: 42 | 43 | fimo_path = fimo_root + cell_line + '_' + data_type + '_all_retest/fimo.txt' 44 | # data_path = data_root + cell_line + '/' + cell_line + '_ep_split.h5' 45 | 46 | matches = dict() # dict mapping motif_names to lists of (sample_idx, start_idx, stop_idx) triples 47 | 48 | print 'Reading and processing FIMO output...' 49 | with open(fimo_path, 'rb') as csv_file: 50 | reader = csv.reader(csv_file, delimiter='\t') 51 | 52 | row_idx = -1 53 | for row in reader: 54 | row_idx += 1 55 | if row_idx == 0: # skip header row 56 | continue 57 | 58 | motif_name = row[0] 59 | if not motif_name in matches: # if this is the first match of that motif 60 | matches[motif_name] = [] 61 | 62 | sample_idx = int(row[1]) 63 | motif_start = int(row[2]) 64 | motif_stop = int(row[3]) 65 | matches[motif_name].append((sample_idx, motif_start, motif_stop)) 66 | 67 | print 'Identified ' + str(len(matches)) + ' distinct motifs.' 68 | 69 | print 'Loading original data...' 70 | # X_enhancers_original, X_promoters_original, y = ld.load_hdf5_ep_split(data_path) 71 | with h5py.File(data_path, 'r') as hf: 72 | X_enhancers_original = np.array(hf.get(cell_line + '_X_enhancers')).transpose((0, 2, 1)) 73 | X_promoters_original = np.array(hf.get(cell_line + '_X_promoters')).transpose((0, 2, 1)) 74 | y = np.array(hf.get(cell_line + 'labels')) 75 | print 'np.shape(X_enhancers_original): ' + str(np.shape(X_enhancers_original)) 76 | print 'np.shape(X_promoters_original): ' + str(np.shape(X_promoters_original)) 77 | print 'np.shape(y): ' + str(np.shape(y)) 78 | 79 | model = bm.build_model(use_JASPAR = False) 80 | 81 | print 'Compiling model...' 82 | opt = Adam(lr = 1e-5) 83 | model.compile(loss = 'binary_crossentropy', 84 | optimizer = opt, 85 | metrics = ["accuracy"]) 86 | print 'Loading ' + cell_line + ' ' + data_type + ' model weights...' 87 | model.load_weights('/home/sss1/Desktop/projects/DeepInteractions/weights/' + cell_line + '-basic.hdf5') 88 | out_root = '/home/sss1/Desktop/projects/DeepInteractions/feature_importances/SPEID/from_HOCOMOCO_motifs/' 89 | out_path = out_root + cell_line + '_' + data_type + '_feature_importance.csv' 90 | 91 | print 'Running predictions on original data' 92 | y_score = model.predict([X_enhancers_original, X_promoters_original], batch_size = 100, verbose = 1) 93 | true_AUPR = average_precision_score(y, y_score) 94 | print 'True AUPR is ' + str(true_AUPR) 95 | true_MS = y_score.mean() 96 | print 'True MS is ' + str(true_MS) 97 | 98 | with open(out_path, 'wb') as csv_file: 99 | 100 | writer = csv.writer(csv_file, delimiter = ',') 101 | writer.writerow(['Motif Name', 'Motif Count', 'AUPR Difference', 'MS Difference']) 102 | for motif, idxs in matches.iteritems(): 103 | print 'Randomizing ' + str(len(idxs)) + ' occurrences of motif ' + motif + ' in ' + cell_line + ' ' + data_type + '...' 104 | if data_type == 'enhancers': 105 | X_enhancers = replace_motifs_in_data(X_enhancers_original, idxs) 106 | X_promoters = X_promoters_original 107 | elif data_type == 'promoters': 108 | X_enhancers = X_enhancers_original 109 | X_promoters = replace_motifs_in_data(X_promoters_original, idxs) 110 | else: 111 | raise ValueError 112 | 113 | print 'Running predictions on motif ' + motif + '...' 114 | y_score = model.predict([X_enhancers, X_promoters], batch_size = 200, verbose = 1) 115 | AUPR = average_precision_score(y, y_score) 116 | print 'AUPR after removing motif ' + motif + ' was ' + str(AUPR) + '\n' 117 | MS = y_score.mean() 118 | print 'MS after removing motif ' + motif + ' was ' + str(MS) + '\n' 119 | 120 | writer.writerow([motif, str(len(idxs)), str(true_AUPR - AUPR), str(true_MS - MS)]) 121 | -------------------------------------------------------------------------------- /pairwise/util.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | import matplotlib.pyplot as plt 3 | from sklearn.metrics import confusion_matrix, log_loss, roc_curve, auc, precision_recall_curve, average_precision_score 4 | 5 | def initialize_with_JASPAR(enhancer_conv_layer, promoter_conv_layer): 6 | JASPAR_motifs = list(np.load('/home/sss1/Desktop/projects/DeepInteractions/JASPAR_CORE_2016_vertebrates.npy')) 7 | print 'Initializing ' + str(len(JASPAR_motifs)) + ' kernels with JASPAR motifs.' 8 | enhancer_conv_weights = enhancer_conv_layer.get_weights() 9 | promoter_conv_weights = promoter_conv_layer.get_weights() 10 | reverse_motifs = [JASPAR_motifs[19][::-1,::-1], JASPAR_motifs[97][::-1,::-1], 11 | JASPAR_motifs[98][::-1,::-1], JASPAR_motifs[99][::-1,::-1], 12 | JASPAR_motifs[100][::-1,::-1], JASPAR_motifs[101][::-1,::-1]] 13 | JASPAR_motifs = JASPAR_motifs + reverse_motifs 14 | for i in xrange(len(JASPAR_motifs)): 15 | m = JASPAR_motifs[i][::-1,:] 16 | w = len(m) 17 | start = np.random.randint(low=3, high=30-w+1-3) 18 | enhancer_conv_weights[0][i,:,start:start+w,0] = m.T - 0.25 19 | enhancer_conv_weights[1][i] = np.random.uniform(low=-1.0,high=0.0) 20 | promoter_conv_weights[0][i,:,start:start+w,0] = m.T - 0.25 21 | promoter_conv_weights[1][i] = np.random.uniform(low=-1.0,high=0.0) 22 | enhancer_conv_layer.set_weights(enhancer_conv_weights) 23 | promoter_conv_layer.set_weights(promoter_conv_weights) 24 | 25 | # Splits the data into training and validation data, keeping training_frac of 26 | # the input samples in the training set and the rest for validation 27 | def split_train_and_val_data(X_enhancer_train, X_promoter_train, y_train, training_frac): 28 | 29 | n_train = int(training_frac * np.shape(y_train)[0]) # number of training samples 30 | 31 | X_enhancer_val = X_enhancer_train[n_train:, :] 32 | X_enhancer_train = X_enhancer_train[:n_train, :] 33 | 34 | X_promoter_val = X_promoter_train[n_train:, :] 35 | X_promoter_train = X_promoter_train[:n_train, :] 36 | 37 | y_val = y_train[n_train:] 38 | y_train = y_train[:n_train] 39 | 40 | return X_enhancer_train, X_promoter_train, y_train, X_enhancer_val, X_promoter_val, y_val 41 | 42 | # Calculates and prints several metrics (confusion matrix, Precision/Recall/F1) 43 | # in real time; also updates the values in the conf_mat_callback so they can be 44 | # plotted or analyzed later 45 | def print_live(conf_mat_callback, y_val, val_predict, logs): 46 | 47 | conf_mat = confusion_matrix(y_val, val_predict).astype(float) 48 | 49 | precision = conf_mat[1, 1] / conf_mat[:, 1].sum() 50 | recall = conf_mat[1, 1] / conf_mat[1, :].sum() 51 | f1_score = 2 * precision * recall / (precision + recall) 52 | 53 | acc = (conf_mat[0, 0] + conf_mat[1, 1]) / np.sum(conf_mat) 54 | 55 | loss = log_loss(y_val, val_predict) 56 | 57 | conf_mat_callback.precisions.append(precision) 58 | conf_mat_callback.recalls.append(recall) 59 | conf_mat_callback.f1_scores.append(f1_score) 60 | conf_mat_callback.losses.append(loss) 61 | conf_mat_callback.accs.append(acc) 62 | print '\nConfusion matrix:\n' + str(conf_mat) + '\n' 63 | print 'Precision: ' + str(precision) + \ 64 | ' Recall: ' + str(recall) + \ 65 | ' F1: ' + str(f1_score) + \ 66 | ' Accuracy: ' + str(acc) + \ 67 | ' Log Loss: ' + str(loss) 68 | print 'Predicted fractions: ' + str(val_predict.mean()) 69 | print 'Actual fractions: ' + str(y_val.mean()) + '\n' 70 | 71 | # Plots several metrics (Precision/Recall/F1, loss, Accuracy) in real time 72 | # (i.e., after each epoch) 73 | def plot_live(conf_mat_callback): 74 | 75 | epoch = conf_mat_callback.epoch 76 | 77 | plt.clf() 78 | xs = [1 + i for i in range(epoch)] 79 | precisions_plot = plt.plot(xs, conf_mat_callback.precisions, label = 'Precision') 80 | recalls_plot = plt.plot(xs, conf_mat_callback.recalls, label = 'Recall') 81 | f1_scores_plot = plt.plot(xs, conf_mat_callback.f1_scores, label = 'F1 score') 82 | accs_plot = plt.plot(xs, conf_mat_callback.accs, label = 'Accuracy') 83 | losses_plot = plt.plot(xs, conf_mat_callback.losses / max(conf_mat_callback.losses), label = 'Loss') 84 | batch_xs = [1 + epoch * float(i)/len(conf_mat_callback.training_losses) for i in range(len(conf_mat_callback.training_losses))] 85 | training_losses_plot = plt.plot(batch_xs, conf_mat_callback.training_losses / max(conf_mat_callback.training_losses), label = 'Training Loss') 86 | training_losses_plot = plt.plot(batch_xs, conf_mat_callback.training_accs, label = 'Training Accuracy') 87 | plt.legend(bbox_to_anchor = (0, 1), loc = 4, borderaxespad = 0., prop={'size':6}) 88 | plt.ylim([0, 1]) 89 | plt.pause(.001) 90 | 91 | # Given a (nearly) balanced data set (i.e., labeled enhancer and promoter 92 | # sequence pairs), subsamples the positive samples to produce the desired 93 | # fraction of positive samples; retains all negative samples 94 | def subsample_imbalanced(X_enhancer, X_promoter, y, positive_subsample_frac): 95 | n = np.shape(y_train)[0] # sample size (i.e., number of pairs) 96 | 97 | # indices that are positive and selected to be retained or negative 98 | to_keep = (np.random(n) < positive_subsample_frac) or (y == 1) 99 | 100 | return X_enhancer[to_keep, :], X_promoter[to_keep, :], y[to_keep] 101 | 102 | 103 | def compute_AUPR(y, y_score): 104 | # print 'Computing Precision-Recall curve...' 105 | precision, recall, _ = precision_recall_curve(y, y_score) 106 | average_precision = average_precision_score(y, y_score) 107 | 108 | def plot_PR_curve(y, y_score): 109 | # print 'Computing Precision-Recall curve...' 110 | precision, recall, _ = precision_recall_curve(y, y_score) 111 | return average_precision_score(y, y_score) 112 | 113 | def plot_ROC_curve(y, y_score): 114 | # print 'Computing ROC curve...' 115 | fpr, tpr, thresholds = roc_curve(y, y_score) 116 | return auc(fpr, tpr) 117 | --------------------------------------------------------------------------------