def writeDAbundanceToFiles(stats, sampleName, outDir, stream=None):
igdDist = Counter(stats["dgene"].tolist())
igdDist = Counter(dict([(str(k), igdDist[k]) for k in igdDist]))
if len(igdDist) == 0:
printto(stream, "WARNING: No IGD hits were detected.", LEVEL.WARN)
return
# Write the counts of all IGVs into a text file
# This isn't plotted by default, but we still write the csv file for it
classes = sorted(igdDist, key=igdDist.get, reverse=True)
total = sum(igdDist.values()) * 1.0
writeCSV(os.path.join(outDir, sampleName + '_igd_dist_variant_level.csv'),
"x,y\n", "{},{}\n", [(x, y) for x, y in zip(
classes, map(lambda k: (igdDist[k] / total * 100), classes))])
# Group IGVs based on the subfamilies (gene level) and then write into a text file
igdDistSub = compressCountsGeneLevel(igdDist)
plotDist(igdDistSub,
sampleName,
os.path.join(outDir, sampleName + '_igd_dist_gene_level.csv'),
rotateLabels=False,
vertical=False,
title='IGD Abundance in Sample ' + sampleName,
stream=stream)
# Group IGVs based on the families and then write into a text file
igdDistfam = compressCountsFamilyLevel(igdDistSub)
# Plot the family level distribution
plotDist(igdDistfam,
sampleName,
os.path.join(outDir, sampleName + '_igd_dist_family_level.csv'),
title='IGD Abundance in Sample ' + sampleName,
stream=stream)
def writeCountsCategoriesToFile(countsVariant, sampleName, filePrefix, title=''):
# gene level
countsVariant = compressCountsGeneLevel(countsVariant)
plotDist(countsVariant, sampleName, filePrefix + 'gene.csv', title)
# family level
countsVariant = compressCountsFamilyLevel(countsVariant)
plotDist(countsVariant, sampleName, filePrefix + 'family.csv', title)
def writeJAbundanceToFiles(stats, sampleName, outDir, stream=None):
igjDist = Counter(stats["jgene"].tolist())
igjDist = dict([(str(k), igjDist[k]) for k in igjDist])
if len(igjDist) == 0:
printto(stream, "WARNING: No IGJ hits were detected.", LEVEL.WARN)
return
plotDist(igjDist,
sampleName,
os.path.join(outDir, sampleName + '_igj_dist_variant_level.csv'),
rotateLabels=False,
vertical=False,
stream=stream)
# Group IGVs based on the subfamilies (gene level) and then write into a text file
igjDistSub = compressCountsGeneLevel(igjDist)
# plotDist(igjDistSub, sampleName, outDir + sampleName +
# '_igj_dist_gene_level.csv', rotateLabels=False, vertical=False)
#
# Group IGVs based on the families and then write into a text file
igjDistfam = compressCountsFamilyLevel(igjDistSub)
# Plot the family level distribution
plotDist(igjDistfam,
sampleName,
os.path.join(outDir, sampleName + '_igj_dist_family_level.csv'),
title='IGJ Abundance in Sample ' + sampleName,
stream=stream)
def writeVAbundanceToFiles(stats, sampleName, outDir, stream=None):
igvDist = Counter(stats["vgene"].tolist())
if len(igvDist) == 0:
printto(stream, "WARNING: No IGV hits were detected.", LEVEL.WARN)
return
# Write the counts of all IGVs into a text file - variant_level isn't plotted by default.
classes = sorted(igvDist, key=igvDist.get, reverse=True)
total = sum(igvDist.values()) * 1.0
writeCSV(os.path.join(outDir, sampleName + '_igv_dist_variant_level.csv'),
"x,y\n", "{},{}\n", [(x, y) for x, y in zip(
classes, map(lambda k: (igvDist[k] / total * 100), classes))])
# Group IGVs based on the subfamilies (gene level) and then write into a text file
igvDistSub = compressCountsGeneLevel(igvDist)
# for k in igvDist.keys():
# ksub = k.split('*')[0]
# igvDistSub[ksub] = igvDistSub.get(ksub, 0) + igvDist[k]
plotDist(igvDistSub,
sampleName,
os.path.join(outDir, sampleName + '_igv_dist_gene_level.csv'),
rotateLabels=False,
vertical=False,
stream=stream)
# Group IGVs based on the families and then write into a text file
igvDistfam = compressCountsFamilyLevel(igvDistSub)
# for k in igvDistSub.keys():
# kfam = k.split('-')[0].split('/')[0]
# igvDistfam[kfam] = igvDistfam.get(kfam, 0) + igvDistSub[k]
# Plot the family level distribution
plotDist(igvDistfam,
sampleName,
os.path.join(outDir, sampleName + '_igv_dist_family_level.csv'),
stream=stream)
# plot alignment length vs %identity
generateStatsHeatmap(
stats,
sampleName, ['alignlen', 'identity'],
['Alignment Length', '%Identity'],
os.path.join(outDir,
sampleName + '_igv_align_quality_identity_hm.tsv'),
stream=stream)
# plot alignment length vs bitScore
generateStatsHeatmap(
stats,
sampleName, ['alignlen', 'bitscore'], ['Alignment Length', 'bitScore'],
os.path.join(outDir,
sampleName + '_igv_align_quality_bitscore_hm.tsv'),
stream=stream)
# plot query start vs. subject start
generateStatsHeatmap(
stats,
sampleName, ['vqstart', 'vstart'], ['Query Start', 'Subject Start'],
os.path.join(outDir, sampleName + '_igv_align_quality_start_hm.tsv'),
stream=stream)
generateStatsHeatmap(
stats,
sampleName, ['alignlen', 'vmismatches'],
['Alignment Length', 'Mismatches'],
os.path.join(outDir,
sampleName + '_igv_align_quality_mismatches_hm.tsv'),
stream=stream)
c = Counter(stats['vmismatches'].tolist())
plotDist(c,
sampleName,
os.path.join(outDir, sampleName + '_igv_mismatches_dist.csv'),
title='Number of Mismatches in V gene',
proportion=True,
rotateLabels=False,
top=20,
stream=stream)
generateStatsHeatmap(
stats,
sampleName, ['alignlen', 'vgaps'], ['Alignment Length', 'Gaps'],
os.path.join(outDir, sampleName + '_igv_align_quality_gaps_hm.tsv'),
stream=stream)
c = Counter(stats['vgaps'].tolist())
plotDist(c,
sampleName,
os.path.join(outDir, sampleName + '_igv_gaps_dist.csv'),
title='Number of Gaps in V gene',
proportion=True,
rotateLabels=False,
top=20,
stream=stream)
def writePrimerStats(end,
name,
cloneAnnot,
fileprefix,
category="All",
stream=None):
NA = str(np.nan)
PRIMER = str(end) + 'endPrimer'
MISMATCH = str(end) + 'endMismatchIndex'
INDEL = str(end) + 'endIndelIndex'
known = cloneAnnot[cloneAnnot[PRIMER] != NA]
integrity = {
'Unknown': (len(cloneAnnot) - len(known)),
'Indelled': sum(known[INDEL] != 0),
'Mismatched': sum(known[MISMATCH] != 0),
'Intact': len(known[(known[INDEL] == 0) & (known[MISMATCH] == 0)])
}
plotDist(integrity,
name,
fileprefix + 'integrity_dist.csv',
title='Integrity of {}\'-end Primer Sequence (%s)'.format(end) %
(category),
proportion=True,
rotateLabels=False)
invalidClones = known.index[known[INDEL] != 0].tolist()
valid = known.index[known[INDEL] == 0].tolist()
printto(
stream,
"Example of Indelled {}'-end: {}".format(end,
str(invalidClones[1:10])),
LEVEL.INFO)
printto(
stream,
"Example of non-indelled {}'-end: {}".format(end, str(valid[1:10])),
LEVEL.INFO)
c1 = Counter(known[known[INDEL] != 0][PRIMER].tolist())
plotDist(c1,
name,
fileprefix + 'indelled_dist.csv',
title='Abundance of Indelled {}\'-end Primers ({})'.format(
end, category),
proportion=False,
rotateLabels=False,
vertical=False,
top=50)
c = Counter(known[known[INDEL] != 0][INDEL].tolist())
plotDist(
c,
name,
fileprefix + 'indel_pos_dist.csv',
title='Abundance of Indel Positions in {}\'-end Primers ({})'.format(
end, category),
proportion=False,
rotateLabels=False,
vertical=True,
sortValues=False,
top=50)
primers = set(known[PRIMER].tolist())
for primer in primers:
# get only ighv abundance of indelled primers
df = known[known[INDEL] != 0]
df = df[df[PRIMER] == primer]
germLineDist = compressCountsGeneLevel(Counter(df['vgene'].tolist()))
plotDist(germLineDist,
name,
fileprefix + primer + '_igv_dist.csv',
title='IGV Abundance of indelled {} ({})'.format(
primer, category),
proportion=False,
vertical=False,
top=20,
rotateLabels=False)