def validateAndSanitizeOptions(self, options):
assertOptionExists(options.runDir, "run directory")
options.runDir = os.path.abspath(options.runDir)
workflowScriptPath = os.path.join(options.runDir,
options.workflowScriptName)
if os.path.exists(workflowScriptPath):
raise OptParseException(
"Run directory already contains workflow script file '%s'. Each analysis must be configured in a separate directory."
% (workflowScriptPath))
# check reference fasta file exists
assertOptionExists(options.referenceFasta, "reference fasta file")
options.referenceFasta = validateFixExistingFileArg(
options.referenceFasta, "reference")
# check for reference fasta index file:
faiFile = options.referenceFasta + ".fai"
if not os.path.isfile(faiFile):
raise OptParseException(
"Can't find expected fasta index file: '%s'" % (faiFile))
# check for bed file of call regions and its index file
options.callRegionsBed = checkFixTabixIndexedFileOption(
options.callRegionsBed, "call-regions bed")
if (options.regionStrList is None) or (len(options.regionStrList)
== 0):
options.genomeRegionList = None
else:
options.genomeRegionList = [
parseGenomeRegion(r) for r in options.regionStrList
]
# validate chromosome names appearing in region tags and callRegions bed file
if (options.callRegionsBed is not None) or (options.genomeRegionList
is not None):
refChromInfo = getFastaInfo(options.referenceFasta)
if options.callRegionsBed is not None:
for chrom in getTabixChromSet(options.tabixBin,
options.callRegionsBed):
if chrom not in refChromInfo:
raise OptParseException(
"Chromosome label '%s', in call regions bed file '%s', not found in reference genome."
% (chrom, options.callRegionsBed))
if options.genomeRegionList is not None:
for (genomeRegionIndex,
genomeRegion) in enumerate(options.genomeRegionList):
chrom = genomeRegion["chrom"]
if chrom not in refChromInfo:
raise OptParseException(
"Chromosome label '%s', parsed from region argument '%s', not found in reference genome."
%
(chrom, options.regionStrList[genomeRegionIndex]))
def getChromIsSkipped(self):
"""
Determine subset of chroms from chromOrder which are completely skipped over
here "skipped" means that not a single base on the chrom is requested for calling or error estimation
\return The set of chromLabels which are skipped
"""
chromIsSkipped = set()
# return empty set when no region selections have been made:
if ((self.params.genomeRegionList is None)
and (self.params.callRegionsBed is None)):
return chromIsSkipped
def allChromosomes():
"""
Return a set of all chromosomes from the reference/alignments in this analysis
"""
return set(self.params.chromOrder)
# first check chromosome coverage of "regions" arguments
if self.params.genomeRegionList is not None:
chromIsSkipped = allChromosomes()
for genomeRegion in self.params.genomeRegionList:
if genomeRegion["chrom"] in chromIsSkipped:
chromIsSkipped.remove(genomeRegion["chrom"])
# further refine coverage based on callRegions BED file
if self.params.callRegionsBed is not None:
callRegionsChroms = getTabixChromSet(self.params.tabixBin,
self.params.callRegionsBed)
chromsNotInCallRegions = allChromosomes() - callRegionsChroms
# Skip the union of:
# 1. chromosomes skipped already due to region arguments
# 2. chromosomes skipped due to callRegions bed track
chromIsSkipped = chromIsSkipped | chromsNotInCallRegions
# if sequencing error estimation is turned on, make sure estimation targets are not skipped:
if self.params.isEstimateSequenceError:
class Constants:
Megabase = 1000000
errorEstimationMinChromSize = self.params.errorEstimationMinChromMb * Megabase
for chrom in self.params.chromSizes:
if self.params.chromSizes[
chrom] < Constants.errorEstimationMinChromSize:
continue
if chrom in chromIsSkipped:
chromIsSkipped.remove(chrom)
return chromIsSkipped
def validateAndSanitizeOptions(self,options) :
assertOptionExists(options.runDir,"run directory")
options.runDir = os.path.abspath(options.runDir)
workflowScriptPath = os.path.join(options.runDir, options.workflowScriptName)
if os.path.exists(workflowScriptPath):
raise OptParseException("Run directory already contains workflow script file '%s'. Each analysis must be configured in a separate directory." % (workflowScriptPath))
# check reference fasta file exists
assertOptionExists(options.referenceFasta,"reference fasta file")
options.referenceFasta=validateFixExistingFileArg(options.referenceFasta,"reference")
# check for reference fasta index file:
faiFile=options.referenceFasta + ".fai"
if not os.path.isfile(faiFile) :
raise OptParseException("Can't find expected fasta index file: '%s'" % (faiFile))
# check for bed file of call regions and its index file
options.callRegionsBed = checkFixTabixIndexedFileOption(options.callRegionsBed, "call-regions bed")
if (options.regionStrList is None) or (len(options.regionStrList) == 0) :
options.genomeRegionList = None
else :
options.genomeRegionList = [parseGenomeRegion(r) for r in options.regionStrList]
# validate chromosome names appearing in region tags and callRegions bed file
if (options.callRegionsBed is not None) or (options.genomeRegionList is not None) :
refChromInfo = getFastaInfo(options.referenceFasta)
if options.callRegionsBed is not None :
for chrom in getTabixChromSet(options.tabixBin, options.callRegionsBed) :
if chrom not in refChromInfo :
raise OptParseException("Chromosome label '%s', in call regions bed file '%s', not found in reference genome." %
(chrom, options.callRegionsBed))
if options.genomeRegionList is not None :
for (genomeRegionIndex, genomeRegion) in enumerate(options.genomeRegionList) :
chrom = genomeRegion["chrom"]
if chrom not in refChromInfo :
raise OptParseException("Chromosome label '%s', parsed from region argument '%s', not found in reference genome." %
(chrom, options.regionStrList[genomeRegionIndex]))
def validateAndSanitizeOptions(self, options):
assertOptionExists(options.runDir, "run directory")
options.runDir = os.path.abspath(options.runDir)
workflowScriptPath = os.path.join(options.runDir,
options.workflowScriptName)
if os.path.exists(workflowScriptPath):
raise OptParseException(
"Run directory already contains workflow script file '%s'. Each analysis must be configured in a separate directory."
% (workflowScriptPath))
assertOptionExists(options.referenceFasta, "reference fasta file")
options.referenceFasta = validateFixExistingFileArg(
options.referenceFasta, "reference fasta file")
# check for reference fasta index file:
referenceFastaIndex = options.referenceFasta + ".fai"
if not os.path.isfile(referenceFastaIndex):
raise OptParseException(
"Can't find expected fasta index file: '%s'" %
(referenceFastaIndex))
if options.isEstimateSequenceError:
# Determine if dynamic error estimation is feasible based on the reference size
# - Given reference contig set (S) with sequence length of at least 5 Mb
# - The total sequence length from S must be at least 50 Mb
class Constants:
Megabase = 1000000
minChromSize = options.errorEstimationMinChromMb * Megabase
minTotalSize = options.errorEstimationMinTotalMb * Megabase
# read fasta index
(_, chromSizes) = getFastaChromOrderSize(referenceFastaIndex)
totalEstimationSize = 0
for chromSize in chromSizes.values():
if chromSize < Constants.minChromSize: continue
totalEstimationSize += chromSize
if totalEstimationSize < Constants.minTotalSize:
sys.stderr.write(
"WARNING: Cannot estimate sequence errors from data due to small or overly fragmented reference sequence. Sequence error estimation disabled.\n"
)
options.isEstimateSequenceError = False
checkFixTabixListOption(options.indelCandidatesList,
"candidate indel vcf")
checkFixTabixListOption(options.forcedGTList, "forced genotype vcf")
options.callRegionsBed = checkFixTabixIndexedFileOption(
options.callRegionsBed, "call-regions bed")
def extendedRegionStrList():
"""
A generator on the regionStrList which parses the (intentionally undocumented/possibly deprecated) '+' entry format
to specify multiple regions in a single argument.
"""
for r in options.regionStrList:
for rr in r.split("+"):
yield rr
if (options.regionStrList is None) or (len(options.regionStrList)
== 0):
options.genomeRegionList = None
else:
options.genomeRegionList = [
parseGenomeRegion(r) for r in extendedRegionStrList()
]
# validate chromosome names appearing in region tags and callRegions bed file
if (options.callRegionsBed is not None) or (options.genomeRegionList
is not None):
refChromInfo = getFastaInfo(options.referenceFasta)
if options.callRegionsBed is not None:
for chrom in getTabixChromSet(options.tabixBin,
options.callRegionsBed):
if chrom not in refChromInfo:
raise OptParseException(
"Chromosome label '%s', in call regions bed file '%s', not found in reference genome."
% (chrom, options.callRegionsBed))
if options.genomeRegionList is not None:
for (genomeRegionIndex,
genomeRegion) in enumerate(options.genomeRegionList):
chrom = genomeRegion["chrom"]
if chrom not in refChromInfo:
raise OptParseException(
"Chromosome label '%s', parsed from region argument '%s', not found in reference genome."
% (chrom, list(
extendedRegionStrList())[genomeRegionIndex]))
options.snvScoringModelFile = validateFixExistingFileArg(
options.snvScoringModelFile, "SNV empirical scoring model file")
options.indelScoringModelFile = validateFixExistingFileArg(
options.indelScoringModelFile,
"Indel empirical scoring model file")
def getCallRegions(params) :
"""
determine
1) a set of genomic regions for calling
2) a set of chromosomes that are completely skipped over,
where "skipped" means that not a single base on the chrom is requested for calling
\return a list of genomic regions for calling
\return a set of chromLabels which are skipped
"""
callRegionList = []
chromIsSkipped = set()
# when no region selections have been made:
if ((params.genomeRegionList is None) and
(params.callRegionsBed is None)) :
return (callRegionList, chromIsSkipped)
# check chromosome coverage of "regions" arguments
chromIsSkipped = set(params.chromOrder)
if params.genomeRegionList is not None :
for genomeRegion in params.genomeRegionList :
chrom = genomeRegion["chrom"]
if chrom not in params.chromOrder:
raise Exception("Unexpected chromosome '%s' in the argument of target regions (--region)" %
(chrom))
if chrom in chromIsSkipped :
chromIsSkipped.remove(chrom)
if params.callRegionsBed is None :
return (params.genomeRegionList, chromIsSkipped)
# check chromsome coverage based on callRegions BED file
callChromList = []
chromIsSkipped2 = set(params.chromOrder)
for chrom in getTabixChromSet(params.tabixBin, params.callRegionsBed) :
if chrom not in params.chromOrder:
raise Exception("Unexpected chromosome '%s' in the bed file of call regions %s " %
(chrom, params.callRegionsBed))
callChromList.append(chrom)
if chrom in chromIsSkipped2 :
chromIsSkipped2.remove(chrom)
if params.genomeRegionList is None :
chromIsSkipped = chromIsSkipped2
for chrom in callChromList:
chromRegion = {"chrom":chrom, "start":1, "end":params.chromSizes[chrom]}
callRegions = getOverlapCallRegions(params, chromRegion)
callRegionList.extend(callRegions)
else:
chromIsSkipped = chromIsSkipped | chromIsSkipped2
for genomeRegion in params.genomeRegionList:
chrom = genomeRegion['chrom']
if genomeRegion["start"] is None:
genomeRegion["start"] = 1
if genomeRegion["end"] is None:
genomeRegion["end"] = params.chromSizes[chrom]
subCallRegions = getOverlapCallRegions(params, genomeRegion)
callRegionList.extend(subCallRegions)
return (callRegionList, chromIsSkipped)
def getCallRegions(params):
"""
determine
1) a set of genomic regions for calling
2) a set of chromosomes that are completely skipped over,
where "skipped" means that not a single base on the chrom is requested for calling
\return a list of genomic regions for calling
\return a set of chromLabels which are skipped
"""
callRegionList = []
chromIsSkipped = set()
# when no region selections have been made:
if ((params.genomeRegionList is None) and (params.callRegionsBed is None)):
return (callRegionList, chromIsSkipped)
# check chromosome coverage of "regions" arguments
chromIsSkipped = set(params.chromOrder)
if params.genomeRegionList is not None:
for genomeRegion in params.genomeRegionList:
chrom = genomeRegion["chrom"]
if chrom not in params.chromOrder:
raise Exception(
"Unexpected chromosome '%s' in the argument of target regions (--region)"
% (chrom))
if chrom in chromIsSkipped:
chromIsSkipped.remove(chrom)
if params.callRegionsBed is None:
return (params.genomeRegionList, chromIsSkipped)
# check chromsome coverage based on callRegions BED file
callChromList = []
chromIsSkipped2 = set(params.chromOrder)
for chrom in getTabixChromSet(params.tabixBin, params.callRegionsBed):
if chrom not in params.chromOrder:
raise Exception(
"Unexpected chromosome '%s' in the bed file of call regions %s "
% (chrom, params.callRegionsBed))
callChromList.append(chrom)
if chrom in chromIsSkipped2:
chromIsSkipped2.remove(chrom)
if params.genomeRegionList is None:
chromIsSkipped = chromIsSkipped2
for chrom in callChromList:
chromRegion = {
"chrom": chrom,
"start": 1,
"end": params.chromSizes[chrom]
}
callRegions = getOverlapCallRegions(params, chromRegion)
callRegionList.extend(callRegions)
else:
chromIsSkipped = chromIsSkipped | chromIsSkipped2
for genomeRegion in params.genomeRegionList:
chrom = genomeRegion['chrom']
if genomeRegion["start"] is None:
genomeRegion["start"] = 1
if genomeRegion["end"] is None:
genomeRegion["end"] = params.chromSizes[chrom]
subCallRegions = getOverlapCallRegions(params, genomeRegion)
callRegionList.extend(subCallRegions)
return (callRegionList, chromIsSkipped)
