def _mainLoop(self):
# Read in the existing modifications.gff
modReader = GffReader(self.args.modifications)
# Set up some additional headers to be injected
headers = [
('source', 'kineticModificationCaller 1.3.1'),
('source-commandline', " ".join(sys.argv)),
('attribute-description', 'modsfwd - count of detected DNA modifications on forward strand'),
('attribute-description', 'modsrev - count of detected DNA modifications on reverse strand')
]
# Get modification calls
hits = [{"pos": x.start, "strand": x.strand} for x in modReader if x.type == 'modified_base']
# Summary reader
summaryFile = file(self.args.alignmentSummary)
# Modified gff file
summaryWriter = file(self.args.outfile, "w")
self.seqMap = {}
inHeader = True
# Loop through
for line in summaryFile:
# Pass any metadata line straight through
if line[0] == "#":
# Parse headers
splitFields = line.replace('#', '').split(' ')
field = splitFields[0]
value = " ".join(splitFields[1:])
if field == 'sequence-header':
[internalTag, delim, externalTag] = value.strip().partition(' ')
self.seqMap[internalTag] = externalTag
print >>summaryWriter, line.strip()
continue
if inHeader:
# We are at the end of the header -- write the tool-specific headers
for field in headers:
print >>summaryWriter, ("##%s %s" % field)
inHeader = False
# Parse the line
rec = Gff3Record.fromString(line)
if rec.type == 'region':
# Get the hits in this interval, add them to the gff record
intervalHits = [h for h in hits if rec.start <= h['pos'] <= rec.end]
strand0Hits = len([h for h in intervalHits if h['strand'] == '+'])
strand1Hits = len([h for h in intervalHits if h['strand'] == '-'])
rec.modsfwd = strand0Hits
rec.modsrev = strand1Hits
print >>summaryWriter, str(rec)
def test_fromString(self):
newRecord = Gff3Record.fromString(str(self.RECORD))
assert str(self.RECORD) == str(newRecord)
def test_fromString(self):
newRecord = Gff3Record.fromString(str(self.record))
assert_equal(str(self.record), str(newRecord))
def main():
headers = [
("source", "GenomicConsensus %s" % __VERSION__),
("pacbio-alignment-summary-version", "0.6"),
("source-commandline", " ".join(sys.argv)),
]
desc = "Augment the alignment_summary.gff file with consensus and variants information."
parser = argparse.ArgumentParser(description=desc)
parser.add_argument("--variantsGff",
type=str,
help="Input variants.gff or variants.gff.gz filename",
required=True)
parser.add_argument("--output",
"-o",
type=str,
help="Output alignment_summary.gff filename")
parser.add_argument("inputAlignmentSummaryGff",
type=str,
help="Input alignment_summary.gff filename")
options = parser.parse_args()
inputVariantsGff = GffReader(options.variantsGff)
inputAlignmentSummaryGff = GffReader(options.inputAlignmentSummaryGff)
summaries = {}
for gffRecord in inputAlignmentSummaryGff:
region = Region(gffRecord.seqid, gffRecord.start, gffRecord.end)
summaries[region] = {"ins": 0, "del": 0, "sub": 0, "cQv": (0, 0, 0)}
inputAlignmentSummaryGff.close()
counterNames = {
"insertion": "ins",
"deletion": "del",
"substitution": "sub"
}
for variantGffRecord in inputVariantsGff:
for region in summaries:
summary = summaries[region]
if (region.seqid == variantGffRecord.seqid
and region.start <= variantGffRecord.start <= region.end):
counterName = counterNames[variantGffRecord.type]
variantLength = max(len(variantGffRecord.reference),
len(variantGffRecord.variantSeq))
summary[counterName] += variantLength
# TODO: base consensusQV on effective coverage
summary["cQv"] = (20, 20, 20)
inputAlignmentSummaryGff = open(options.inputAlignmentSummaryGff)
outputAlignmentSummaryGff = open(options.output, "w")
inHeader = True
for line in inputAlignmentSummaryGff:
line = line.rstrip()
# Pass any metadata line straight through
if line[0] == "#":
print >> outputAlignmentSummaryGff, line.strip()
continue
if inHeader:
# We are at the end of the header -- write the tool-specific headers
for k, v in headers:
print >> outputAlignmentSummaryGff, ("##%s %s" % (k, v))
inHeader = False
# Parse the line
rec = Gff3Record.fromString(line)
if rec.type == "region":
summary = summaries[(rec.seqid, rec.start, rec.end)]
if "cQv" in summary:
cQvTuple = summary["cQv"]
line += ";%s=%s" % ("cQv", ",".join(
str(int(f)) for f in cQvTuple))
for counterName in counterNames.values():
if counterName in summary:
line += ";%s=%d" % (counterName, summary[counterName])
print >> outputAlignmentSummaryGff, line
def _mainLoop(self):
# Read in the existing modifications.gff
modReader = GffReader(self.modifications)
headerString = ",".join(['"' + x + '"' for x in self.knownModificationEvents])
# Set up some additional headers to be injected
headers = [
("source", "kineticModificationCaller 1.3.3"),
("source-commandline", " ".join(sys.argv)),
(
"attribute-description",
"modsfwd - count of detected DNA modifications on forward strand by modification event type",
),
(
"attribute-description",
"modsrev - count of detected DNA modifications on reverse strand by modification event type",
),
("region-modsfwd", headerString),
("region-modsfwd", headerString),
]
hitsByEvent = dict([(x, []) for x in self.knownModificationEvents])
# Get modification calls
hits = [
{"pos": x.start, "strand": x.strand, "seqid": x.seqid, "type": x.type}
for x in modReader
if x.type in self.knownModificationEvents
]
# Summary reader
summaryFile = file(self.alignmentSummary)
# Modified gff file
summaryWriter = file(self.outfile, "w")
self.seqMap = {}
inHeader = True
# Loop through
for line in summaryFile:
# Pass any metadata line straight through
if line[0] == "#":
# Parse headers
splitFields = line.replace("#", "").split(" ")
field = splitFields[0]
value = " ".join(splitFields[1:])
if field == "sequence-header":
[internalTag, delim, externalTag] = value.strip().partition(" ")
self.seqMap[internalTag] = externalTag
print >> summaryWriter, line.strip()
continue
if inHeader:
# We are at the end of the header -- write the tool-specific headers
for field in headers:
print >> summaryWriter, ("##%s %s" % field)
inHeader = False
# Parse the line
rec = Gff3Record.fromString(line)
if rec.type == "region":
# Get the hits in this interval, add them to the gff record
intervalHits = [h for h in hits if rec.start <= h["pos"] <= rec.end and rec.seqid == h["seqid"]]
cFwd = self.countModificationTypes([h for h in intervalHits if h["strand"] == "+"])
cRev = self.countModificationTypes([h for h in intervalHits if h["strand"] == "-"])
rec.modsfwd = ",".join([str(cFwd[x]) for x in self.knownModificationEvents])
rec.modsrev = ",".join([str(cRev[x]) for x in self.knownModificationEvents])
print >> summaryWriter, str(rec)
return 0
def test_fromString(self):
newRecord = Gff3Record.fromString(str(self.record))
assert_equal(str(self.record), str(newRecord))
def _mainLoop(self):
# Read in the existing modifications.gff
modReader = GffReader(self.modifications)
headerString = ",".join(
['"' + x + '"' for x in self.knownModificationEvents])
# Set up some additional headers to be injected
headers = [
('source', 'kineticModificationCaller 1.3.3'),
('source-commandline', " ".join(sys.argv)),
('attribute-description',
'modsfwd - count of detected DNA modifications on forward strand by modification event type'
),
('attribute-description',
'modsrev - count of detected DNA modifications on reverse strand by modification event type'
), ('region-modsfwd', headerString),
('region-modsfwd', headerString)
]
hitsByEvent = dict([(x, []) for x in self.knownModificationEvents])
# Get modification calls
hits = [{
"pos": x.start,
"strand": x.strand,
"seqid": x.seqid,
"type": x.type
} for x in modReader if x.type in self.knownModificationEvents]
# Summary reader
summaryFile = file(self.alignmentSummary)
# Modified gff file
summaryWriter = file(self.outfile, "w")
self.seqMap = {}
inHeader = True
# Loop through
for line in summaryFile:
# Pass any metadata line straight through
if line[0] == "#":
# Parse headers
splitFields = line.replace('#', '').split(' ')
field = splitFields[0]
value = " ".join(splitFields[1:])
if field == 'sequence-header':
[internalTag, delim,
externalTag] = value.strip().partition(' ')
self.seqMap[internalTag] = externalTag
print(line.strip(), file=summaryWriter)
continue
if inHeader:
# We are at the end of the header -- write the tool-specific headers
for field in headers:
print(("##%s %s" % field), file=summaryWriter)
inHeader = False
# Parse the line
rec = Gff3Record.fromString(line)
if rec.type == 'region':
# Get the hits in this interval, add them to the gff record
intervalHits = [
h for h in hits if rec.start <= h['pos'] <= rec.end
and rec.seqid == h['seqid']
]
cFwd = self.countModificationTypes(
[h for h in intervalHits if h['strand'] == '+'])
cRev = self.countModificationTypes(
[h for h in intervalHits if h['strand'] == '-'])
rec.modsfwd = ",".join(
[str(cFwd[x]) for x in self.knownModificationEvents])
rec.modsrev = ",".join(
[str(cRev[x]) for x in self.knownModificationEvents])
print(str(rec), file=summaryWriter)
return 0
