This file is indexed.

/usr/lib/python2.7/dist-packages/GenomicConsensus/options.py is in python-pbgenomicconsensus 2.1.0-1.

This file is owned by root:root, with mode 0o644.

The actual contents of the file can be viewed below.

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
#################################################################################
# Copyright (c) 2011-2013, Pacific Biosciences of California, Inc.
#
# All rights reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are met:
# * Redistributions of source code must retain the above copyright
#   notice, this list of conditions and the following disclaimer.
# * Redistributions in binary form must reproduce the above copyright
#   notice, this list of conditions and the following disclaimer in the
#   documentation and/or other materials provided with the distribution.
# * Neither the name of Pacific Biosciences nor the names of its
#   contributors may be used to endorse or promote products derived from
#   this software without specific prior written permission.
#
# NO EXPRESS OR IMPLIED LICENSES TO ANY PARTY'S PATENT RIGHTS ARE GRANTED BY
# THIS LICENSE.  THIS SOFTWARE IS PROVIDED BY PACIFIC BIOSCIENCES AND ITS
# CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
# PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL PACIFIC BIOSCIENCES OR
# ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
# EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
# PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR
# BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER
# IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
# ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
# POSSIBILITY OF SUCH DAMAGE.
#################################################################################

# Author: David Alexander

#
# This module makes the options globally available to all processes.
#
# Presently it relies on the fact that multiprocessing on Linux/Unix uses fork(),
# so the child processes inherit state from the main process.  If we ever wanted
# to port to Windows, we could support this module using the Manager protocol in
# in multiprocessing.
#
# Usage:
#  In the main process, before forking:
#    > from options import parseOptions, options
#    > parseOptions()
#    ...
#  then in any subprocess you can say
#    > from options import options
#  and get the loaded options dictionary.
#
from __future__ import absolute_import
import argparse, h5py, os, os.path, sys, json

from pbcommand.models import FileTypes, SymbolTypes, get_pbparser
from pbcommand.common_options import (add_resolved_tool_contract_option,
                                      add_debug_option)
# FIXME                               add_subcomponent_versions_option)

from .utils import fileFormat
from . import __VERSION__

options = argparse.Namespace()

def consensusCoreVersion():
    try:
        import ConsensusCore
        return ConsensusCore.Version.VersionString()
    except:
        return None

def consensusCore2Version():
    try:
        import ConsensusCore2
        return ConsensusCore2.__version__
    except:
        return None

class Constants(object):
    TOOL_ID = "genomic_consensus.tasks.variantcaller"
    DRIVER_EXE = "variantCaller --resolved-tool-contract "
    ALGORITHM_ID = "genomic_consensus.task_options.algorithm"
    MIN_CONFIDENCE_ID = "genomic_consensus.task_options.min_confidence"
    MIN_COVERAGE_ID = "genomic_consensus.task_options.min_coverage"
    DIPLOID_MODE_ID = "genomic_consensus.task_options.diploid"

    DEFAULT_ALGORITHM = "best"
    DEFAULT_MIN_CONFIDENCE = 40
    DEFAULT_MIN_COVERAGE = 5
    DEFAULT_MAX_COVERAGE = 100
    DEFAULT_MIN_MAPQV = 10
    DEFAULT_MIN_READSCORE = 0.65
    DEFAULT_MIN_HQREGIONSNR = 3.75
    DEFAULT_MIN_ZSCORE = -3.5
    DEFAULT_MIN_ACCURACY = 0.82

def get_parser():
    """
    Construct a hybrid PbParser with most tool contract parameters defined
    separately from argparser parameters.
    """
    p = get_pbparser(
        tool_id=Constants.TOOL_ID,
        version=__VERSION__,
        name="variantCaller",
        description="Compute genomic consensus and call variants relative to the reference.",
        driver_exe=Constants.DRIVER_EXE,
        nproc=SymbolTypes.MAX_NPROC,
        resource_types=(),
        default_level="WARN")
    tcp = p.tool_contract_parser
    tcp.add_input_file_type(FileTypes.DS_ALIGN, "infile",
        "Alignment DataSet", "BAM or Alignment DataSet")
    tcp.add_input_file_type(FileTypes.DS_REF, "reference",
        "Reference DataSet", "Fasta or Reference DataSet")
    tcp.add_output_file_type(FileTypes.GFF, "variants",
        name="Consensus GFF",
        description="Consensus GFF",
        default_name="variants")
    tcp.add_output_file_type(FileTypes.DS_CONTIG, "consensus",
        name="Consensus ContigSet",
        description="Consensus sequence in Fasta format",
        default_name="consensus")
    tcp.add_output_file_type(FileTypes.FASTQ, "consensus_fastq",
        name="Consensus fastq",
        description="Consensus fastq",
        default_name="consensus")
    tcp.add_str(
        option_id=Constants.ALGORITHM_ID,
        option_str="algorithm",
        default=Constants.DEFAULT_ALGORITHM,
        name="Algorithm",
        description="Variant calling algorithm")
    tcp.add_int(
        option_id=Constants.MIN_CONFIDENCE_ID,
        option_str="minConfidence",
        default=Constants.DEFAULT_MIN_CONFIDENCE,
        name="Minimum confidence",
        description="The minimum confidence for a variant call to be output "+\
                    "to variants.gff")
    tcp.add_int(
        option_id=Constants.MIN_COVERAGE_ID,
        option_str="minCoverage",
        default=Constants.DEFAULT_MIN_COVERAGE,
        name="Minimum coverage",
        description="The minimum site coverage that must be achieved for " +\
                    "variant calls and consensus to be calculated for a site.")

    tcp.add_boolean(
        option_id=Constants.DIPLOID_MODE_ID,
        option_str="diploid",
        default=False,
        name="Diploid mode (experimental)",
        description="Enable detection of heterozygous variants (experimental)")
    add_options_to_argument_parser(p.arg_parser.parser)
    return p

def add_options_to_argument_parser(parser):

    def canonicalizedFilePath(path):
        return os.path.abspath(os.path.expanduser(path))

    basics = parser.add_argument_group("Basic required options")
    basics.add_argument(
        "inputFilename",
        type=canonicalizedFilePath,
        help="The input cmp.h5 file")
    basics.add_argument(
        "--referenceFilename", "--reference", "-r",
        action="store",
        dest="referenceFilename",
        type=canonicalizedFilePath,
        required=True,
        help="The filename of the reference FASTA file")
    basics.add_argument(
        "-o", "--outputFilename",
        dest="outputFilenames",
        required=True,
        type=str,
        action="append",
        default=[],
        help="The output filename(s), as a comma-separated list." + \
             "Valid output formats are .fa/.fasta, .fq/.fastq, .gff")

    parallelism = parser.add_argument_group("Parallelism")
    parallelism.add_argument(
        "-j", "--numWorkers",
        dest="numWorkers",
        type=int,
        default=1,
        help="The number of worker processes to be used")

    filtering = parser.add_argument_group("Output filtering")
    filtering.add_argument(
        "--minConfidence", "-q",
        action="store",
        dest="minConfidence",
        type=int,
        default=Constants.DEFAULT_MIN_CONFIDENCE,
        help="The minimum confidence for a variant call to be output to variants.gff")
    filtering.add_argument(
        "--minCoverage", "-x",
        action="store",
        dest="minCoverage",
        default=Constants.DEFAULT_MIN_COVERAGE,
        type=int,
        help="The minimum site coverage that must be achieved for variant calls and " + \
             "consensus to be calculated for a site.")
    filtering.add_argument(
        "--noEvidenceConsensusCall",
        action="store",
        choices=["nocall", "reference", "lowercasereference"],
        default="lowercasereference",
        help="The consensus base that will be output for sites with no effective coverage.")


    readSelection = parser.add_argument_group("Read selection/filtering")
    readSelection.add_argument(
        "--coverage", "-X",
        action="store",
        dest="coverage",
        type=int,
        default=Constants.DEFAULT_MAX_COVERAGE,
        help="A designation of the maximum coverage level to be used for analysis." + \
             " Exact interpretation is algorithm-specific.")
    readSelection.add_argument(
        "--minMapQV", "-m",
        action="store",
        dest="minMapQV",
        type=float,
        default=Constants.DEFAULT_MIN_MAPQV,
        help="The minimum MapQV for reads that will be used for analysis.")
    # Since the reference isn't loaded at options processing time, we
    # can't grok the referenceWindow specified until later.  We store
    # it as a string (referenceWindowsAsString) and it will later be
    # interpreted and stored as a proper window tuple (referenceWindow)
    readSelection.add_argument(
        "--referenceWindow", "--referenceWindows", "-w",
        action="store",
        dest="referenceWindowsAsString",
        type=str,
        help="The window (or multiple comma-delimited windows) of the reference to " + \
             "be processed, in the format refGroup:refStart-refEnd "                 + \
             "(default: entire reference).",
        default=None)

    readSelection.add_argument(
        "--alignmentSetRefWindows",
        action="store_true",
        dest="referenceWindowsFromAlignment",
        help="The window (or multiple comma-delimited windows) of the reference to " + \
             "be processed, in the format refGroup:refStart-refEnd "                 + \
             "will be pulled from the alignment file.",
        default=False)

    def slurpWindowFile(fname):
        return ",".join(map(str.strip, open(fname).readlines()))

    readSelection.add_argument(
        "--referenceWindowsFile", "-W",
        action="store",
        dest="referenceWindowsAsString",
        type=slurpWindowFile,
        help="A file containing reference window designations, one per line",
        default=None)
    readSelection.add_argument(
        "--barcode",
        type=str,
        dest="_barcode",
        help="Only process reads with the given barcode name.")
    def parseReadStratum(s):
        rs = map(int, s.split("/"))
        assert len(rs) == 2
        assert rs[0] < rs[1]
        return rs
    readSelection.add_argument(
        "--readStratum",
        help="A string of the form 'n/N', where n, and N are integers, 0 <= n < N, designating" \
             " that the reads are to be deterministically split into N strata of roughly even"  \
             " size, and stratum n is to be used for variant and consensus calling.  This is"   \
             " mostly useful for Quiver development.",
        dest="readStratum",
        default=None,
        type=parseReadStratum)
    readSelection.add_argument(
        "--minReadScore",
        action="store",
        dest="minReadScore",
        type=float,
        default=Constants.DEFAULT_MIN_READSCORE,
        help="The minimum ReadScore for reads that will be used for analysis (arrow-only).")
    readSelection.add_argument(
        "--minSnr",
        action="store",
        dest="minHqRegionSnr",
        type=float,
        default=Constants.DEFAULT_MIN_HQREGIONSNR,
        help="The minimum acceptable signal-to-noise over all channels for reads that will be used for analysis (arrow-only).")
    readSelection.add_argument(
        "--minZScore",
        action="store",
        dest="minZScore",
        type=float,
        default=Constants.DEFAULT_MIN_ZSCORE,
        help="The minimum acceptable z-score for reads that will be used for analysis (arrow-only).")
    readSelection.add_argument(
        "--minAccuracy",
        action="store",
        dest="minAccuracy",
        type=float,
        default=Constants.DEFAULT_MIN_ACCURACY,
        help="The minimum acceptable window-global alignment accuracy for reads that will be used for the analysis (arrow-only).")

    algorithm = parser.add_argument_group("Algorithm and parameter settings")
    algorithm.add_argument(
        "--algorithm",
        action="store",
        dest="algorithm",
        type=str,
        choices=["quiver", "arrow", "plurality", "poa", "best"],
        default="best")
    algorithm.add_argument(
        "--parametersFile", "-P",
        dest="parametersFile",
        type=str,
        default=None,
        help="Parameter set filename (QuiverParameters.ini), or directory D " + \
             "such that either D/*/GenomicConsensus/QuiverParameters.ini, "   + \
             "or D/GenomicConsensus/QuiverParameters.ini, is found.  In the " + \
             "former case, the lexically largest path is chosen.")
    algorithm.add_argument(
        "--parametersSpec", "-p",
        action="store",
        dest="parametersSpec",
        type=str,
        default="auto",
        help="Name of parameter set (chemistry.model) to select from the "   + \
             "parameters file, or just the name of the chemistry, in which " + \
             "case the best available model is chosen.  Default is 'auto', " + \
             "which selects the best parameter set from the cmp.h5")

    debugging = parser.add_argument_group("Verbosity and debugging/profiling")
    add_debug_option(debugging)
    debugging.add_argument(
        "--pdbAtStartup",
        action="store_true",
        dest="pdbAtStartup",
        default=False,
        help="Drop into Python debugger at startup (requires ipdb)")
    debugging.add_argument(
        "--profile",
        action="store_true",
        dest="doProfiling",
        default=False,
        help="Enable Python-level profiling (using cProfile).")
    debugging.add_argument(
        "--dumpEvidence", "-d",
        dest="dumpEvidence",
        nargs="?",
        default=None,
        const="variants",
        choices=["variants", "all", "outliers"])
    debugging.add_argument(
        "--evidenceDirectory",
        default="evidence_dump")
    debugging.add_argument(
        "--annotateGFF",
        action="store_true",
        help="Augment GFF variant records with additional information")

    advanced = parser.add_argument_group("Advanced configuration options")
    advanced.add_argument(
        "--diploid",
        action="store_true",
        help="Enable detection of heterozygous variants (experimental)")
    advanced.add_argument(
        "--queueSize", "-Q",
        action="store",
        dest="queueSize",
        type=int,
        default=200)
    advanced.add_argument(
        "--threaded", "-T",
        action="store_true",
        dest="threaded",
        default=False,
        help="Run threads instead of processes (for debugging purposes only)")
    advanced.add_argument(
        "--referenceChunkSize", "-C",
        action="store",
        dest="referenceChunkSize",
        type=int,
        default=500)
    advanced.add_argument(
        "--fancyChunking",
        default=True,
        action="store_true",
        help="Adaptive reference chunking designed to handle coverage cutouts better")
    advanced.add_argument(
        "--simpleChunking",
        dest="fancyChunking",
        action="store_false",
        help="Disable adaptive reference chunking")
    advanced.add_argument(
        "--referenceChunkOverlap",
        action="store",
        dest="referenceChunkOverlap",
        type=int,
        default=5)
    advanced.add_argument(
        "--autoDisableHdf5ChunkCache",
        action="store",
        type=int,
        default=500,
        help="Disable the HDF5 chunk cache when the number of datasets in the cmp.h5 " + \
             "exceeds the given threshold")
    advanced.add_argument(
        "--aligner", "-a",
        action="store",
        choices=["affine", "simple"],
        default="affine",
        help="The pairwise alignment algorithm that will be used to produce variant calls" \
             " from the consensus (Quiver only).")
    advanced.add_argument(
        "--refineDinucleotideRepeats",
        dest="refineDinucleotideRepeats",
        action="store_true",
        help="Require quiver maximum likelihood search to try one less/more repeat copy in"  \
             " dinucleotide repeats, which seem to be the most frequent cause of suboptimal" \
             " convergence (getting trapped in local optimum) (Quiver only)")
    advanced.add_argument(
        "--noRefineDinucleotideRepeats",
        dest="refineDinucleotideRepeats",
        action="store_false",
        help="Disable dinucleotide refinement")
    advanced.set_defaults(refineDinucleotideRepeats=True)
    advanced.add_argument(
        "--fast",
        dest="fastMode",
        action="store_true",
        help="Cut some corners to run faster.  Unsupported!")
    advanced.add_argument(
        "--skipUnrecognizedContigs",
        action="store_true",
        help="Do not abort when told to process a reference window (via -w/--referenceWindow[s]) " \
             "that has no aligned coverage.  Outputs emptyish files if there are no remaining "    \
             "non-degenerate windows.  Only intended for use by smrtpipe scatter/gather.")

    return parser

def processOptions():
    """
    Various additions to the global 'options' object, assuming that the
    command-line arguments have already been processed.
    """
    parser = get_parser().arg_parser.parser
    def checkInputFile(path):
        if not os.path.isfile(path):
            parser.error("Input file %s not found." % (path,))

    def checkOutputFile(path):
        try:
            f = open(path, "a")
            f.close()
        except:
            parser.error("Output file %s cannot be written." % (path,))

    options.gffOutputFilename   = None
    options.fastaOutputFilename = None
    options.fastqOutputFilename = None
    options.csvOutputFilename   = None


    for outputFilename in options.outputFilenames:
        fmt = fileFormat(outputFilename)
        if   fmt == "GFF":   options.gffOutputFilename   = outputFilename
        elif fmt == "FASTA": options.fastaOutputFilename = outputFilename
        elif fmt == "FASTQ": options.fastqOutputFilename = outputFilename
        elif fmt == "CSV":   options.csvOutputFilename   = outputFilename

    if options.inputFilename.endswith(".bam"):
        options.usingBam, options.usingCmpH5 = True, False
    else:
        options.usingBam, options.usingCmpH5 = False, True

    for path in (options.inputFilename, options.referenceFilename):
        if path != None:
            checkInputFile(path)

    for path in options.outputFilenames:
        if path != None:
            checkOutputFile(path)

    options.shellCommand = " ".join(sys.argv)


def resolveOptions(alnFile):
    """
    Some of the options are provided as strings by the user, but need
    to be translated into internal identifiers.  These options are
    encoded as options._optionName; here we lookup the ID and store it
    as options.optionName.

    This is essentially just an order-of-initialization issue.
    """
    if options._barcode != None:
        if not alnFile.isBarcoded:
            raise Exception("input file is not barcoded!")
        if options._barcode not in alnFile.barcode:
            raise Exception("Barcode with given name not present in input file!")
        options.barcode = alnFile.barcode[options._barcode]
    else:
        options.barcode = None