def bam_find_regions(bam_name, merge_distance=10, min_read_count=2, only_uniq_starts=False, nostrand=False, out=sys.stdout):
bamfile = bam_open(bam_name)
region_plus = None
region_minus = None
for pileup in bam_pileup_iter(bamfile, mask=1540):
chrom = bamfile.getrname(pileup.tid)
for read in pileup.pileups:
if read.is_del:
continue
if nostrand or not read.alignment.is_reverse:
if not region_plus or region_plus.chrom != chrom or (region_plus.end + merge_distance) < pileup.pos:
if region_plus and region_plus.read_count >= min_read_count:
region_plus.write(out)
region_plus = ExpressedRegion(chrom, only_uniq_starts)
region_plus.add_column(read, pileup.pos)
else:
if not region_minus or region_minus.chrom != chrom or (region_minus.end + merge_distance) < pileup.pos:
if region_minus and region_minus.read_count >= min_read_count:
region_minus.write(out)
region_minus = ExpressedRegion(chrom, only_uniq_starts)
region_minus.add_column(read, pileup.pos)
if region_plus and region_plus.read_count >= min_read_count:
region_plus.write(out)
if region_minus and region_minus.read_count >= min_read_count:
region_minus.write(out)
bamfile.close()
def bam_tofastx(fname,
colorspace=False,
show_mapped=True,
show_unmapped=True,
fastq=True,
read1=True,
read2=True,
proper=False):
if show_mapped is False and show_unmapped is False:
return
sam = bam_open(fname)
last_key = None
for read in bam_iter(sam):
if not read1 and read.is_read1:
continue
if not read2 and read.is_read2:
continue
if proper and not read.is_proper_pair:
continue
k = (read.qname, read.seq)
if last_key == k:
continue
show = False
if show_mapped and not read.is_unmapped:
show = True
if show_unmapped and read.is_unmapped:
show = True
if not show:
continue
if fastq:
write_fastq(read, colorspace=colorspace)
else:
write_fasta(read, colorspace=colorspace)
last_key = k
def bam_find_regions(bam_name,
merge_distance=10,
min_read_count=2,
only_uniq_starts=False,
nostrand=False,
out=sys.stdout):
bamfile = bam_open(bam_name)
region_plus = None
region_minus = None
for pileup in bam_pileup_iter(bamfile, mask=1540):
chrom = bamfile.getrname(pileup.tid)
for read in pileup.pileups:
if read.is_del:
continue
if nostrand or not read.alignment.is_reverse:
if not region_plus or region_plus.chrom != chrom or (
region_plus.end + merge_distance) < pileup.pos:
if region_plus and region_plus.read_count >= min_read_count:
region_plus.write(out)
region_plus = ExpressedRegion(chrom, only_uniq_starts)
region_plus.add_column(read, pileup.pos)
else:
if not region_minus or region_minus.chrom != chrom or (
region_minus.end + merge_distance) < pileup.pos:
if region_minus and region_minus.read_count >= min_read_count:
region_minus.write(out)
region_minus = ExpressedRegion(chrom, only_uniq_starts)
region_minus.add_column(read, pileup.pos)
if region_plus and region_plus.read_count >= min_read_count:
region_plus.write(out)
if region_minus and region_minus.read_count >= min_read_count:
region_minus.write(out)
bamfile.close()
def bam_tofastx(fname, colorspace=False, show_mapped=True, show_unmapped=True, fastq=True, read1=True, read2=True, proper=False):
if show_mapped is False and show_unmapped is False:
return
sam = bam_open(fname)
last_key = None
for read in bam_iter(sam):
if not read1 and read.is_read1:
continue
if not read2 and read.is_read2:
continue
if proper and not read.is_proper_pair:
continue
k = (read.qname, read.seq)
if last_key == k:
continue
show = False
if show_mapped and not read.is_unmapped:
show = True
if show_unmapped and read.is_unmapped:
show = True
if not show:
continue
if fastq:
write_fastq(read, colorspace=colorspace)
else:
write_fasta(read, colorspace=colorspace)
last_key = k
def usage():
print __doc__
print """\
Usage: bamutils peakheight {options} bamfile peaks.bed
"""
sys.exit(1)
if __name__ == "__main__":
bam_fname = None
bed_fname = None
for arg in sys.argv[1:]:
if arg == '-h':
usage()
elif not bam_fname and os.path.exists(arg):
bam_fname = arg
elif not bed_fname and os.path.exists(arg):
bed_fname = arg
else:
print 'Unknown argument: %s' % arg
usage()
if not bam_fname or not bed_fname:
usage()
bam = bam_open(bam_fname)
with open(bed_fname) as f:
bam_peakheight(bam, f)
bam.close()
def bam_stats(infiles, gtf_file=None, region=None, delim=None, tags=[], show_all=False, fillin_stats=True):
if gtf_file:
gtf = GTF(gtf_file)
else:
gtf = None
sys.stderr.write('Calculating Read stats...\n')
stats = [BamStats(bam_open(x), gtf, region, delim, tags, show_all=show_all) for x in infiles]
sys.stdout.write('\t')
for fname, stat in zip(infiles, stats):
sys.stdout.write('%s\t\t' % fname)
sys.stdout.write('\n')
sys.stdout.write('Reads:\t')
for stat in stats:
sys.stdout.write('%s\t\t' % stat.total)
sys.stdout.write('\n')
sys.stdout.write('Mapped:\t')
for stat in stats:
sys.stdout.write('%s\t\t' % stat.mapped)
sys.stdout.write('\n')
sys.stdout.write('Unmapped:\t')
for stat in stats:
sys.stdout.write('%s\t\t' % stat.unmapped)
sys.stdout.write('\n')
sys.stdout.write('\nFlag distribution\n')
validflags = set()
maxsize = 0
for flag in flag_descriptions:
for stat in stats:
if stat.flag_counts.counts[flag] > 0:
validflags.add(flag)
maxsize = max(maxsize, len(flag_descriptions[flag]))
for flag in sorted(validflags):
sys.stdout.write("[0x%03x] %-*s" % (flag, maxsize, flag_descriptions[flag]))
for stat in stats:
sys.stdout.write('\t%s\t%0.2f%%' % (stat.flag_counts.counts[flag], (float(stat.flag_counts.counts[flag]) * 100 / stat.total)))
sys.stdout.write('\n')
sys.stdout.write('\n')
if stats[0].tlen_counts:
sys.stdout.write('Template length:')
for stat in stats:
mean, stdev = counts_mean_stdev(stat.tlen_counts)
sys.stdout.write('\t%0.2f\t+/- %0.2f' % (mean, stdev))
sys.stdout.write('\n')
sys.stdout.write('\n')
stat_tags = {}
for tag in stats[0].tagbins:
stat_tags[tag] = []
for stat in stats:
stat_tags[tag].append(stat.tagbins[tag])
for tag in stat_tags:
asc = stats[0].tagbins[tag].asc
sys.stdout.write("Ave %s:" % tag)
for i, tagbin in enumerate(stat_tags[tag]):
sys.stdout.write('\t%s' % tagbin.mean)
if i != len(stats):
sys.stdout.write('\t')
sys.stdout.write('\n')
sys.stdout.write("Max %s:" % tag)
for i, tagbin in enumerate(stat_tags[tag]):
sys.stdout.write('\t%s' % tagbin.max)
if i != len(stats):
sys.stdout.write('\t')
sys.stdout.write('\n')
sys.stdout.write('%s distribution:\n' % tag)
gens = []
gen_vals = []
last_pcts = []
for stat in stats:
gens.append(stat.distribution_gen(tag))
gen_vals.append(None)
last_pcts.append(0.0)
good = True
last = None
while good:
good = False
for i, stat in enumerate(stats):
if not gen_vals[i]:
try:
gen_vals[i] = gens[i].next()
except StopIteration:
pass
vals = [tup[0] for tup in gen_vals if tup]
if not vals:
continue
if asc:
minval = min(vals)
else:
minval = max(vals)
if last and type(last) == int and fillin_stats:
if asc:
last += 1
# fill in missing values
while last < minval:
sys.stdout.write('%s' % last)
for i, stat in enumerate(stats):
sys.stdout.write('\t0\t%s' % last_pcts[i])
sys.stdout.write('\n')
last += 1
else:
last -= 1
# fill in missing values
while last > minval:
sys.stdout.write('%s' % last)
for i, stat in enumerate(stats):
sys.stdout.write('\t0\t%s' % last_pcts[i])
sys.stdout.write('\n')
last -= 1
last = minval
sys.stdout.write(str(minval))
for i, tup in enumerate(gen_vals):
if tup and tup[0] == minval:
sys.stdout.write('\t%s\t%s' % (tup[1], tup[2]))
last_pcts[i] = tup[2]
gen_vals[i] = None
good = True
else:
sys.stdout.write('\t0\t%s' % (last_pcts[i]))
sys.stdout.write('\n')
sys.stdout.write('\n')
sys.stdout.write('Reference counts')
for stat in stats:
sys.stdout.write('\tcount\t')
sys.stdout.write('\n')
for k in sorted([x for x in stats[0].refs]):
sys.stdout.write('%s' % k)
for stat in stats:
sys.stdout.write('\t%s\t' % stat.refs[k])
sys.stdout.write('\n')
if gtf_file:
sys.stdout.write('Mapping regions')
for stat in stats:
sys.stdout.write('\tcount\tCPM')
sys.stdout.write('\n')
sorted_keys = [x for x in stats[0].regiontagger.counts]
sorted_keys.sort()
for k in sorted_keys:
sys.stdout.write('%s' % k)
for stat in stats:
sys.stdout.write('\t%s\t%s' % (stat.regiontagger.counts[k], float(stat.regiontagger.counts[k]) / stat.mapped / 1000000))
sys.stdout.write('\n')
usage()
elif not ref and os.path.exists(arg) and os.path.exists('%s.fai' % arg):
if os.path.exists('%s.fai' % arg):
ref = arg
else:
print "Missing FAI index on %s" % arg
usage()
elif not regions:
regions = BedFile(region=arg)
else:
print "Unknown option or missing index: %s" % arg
usage()
except Exception, e:
print e
usage()
if not bam:
usage()
else:
bamobj = bam_open(bam)
if profile:
import cProfile
def func():
bam_basecall(bamobj, ref, min_qual, min_count, regions, mask, quiet, showgaps, showstrand, minorpct, altfreq, variants, TimedProfiler())
sys.stderr.write('Profiling...\n')
cProfile.run('func()', profile)
else:
bam_basecall(bamobj, ref, min_qual, min_count, regions, mask, quiet, showgaps, showstrand, minorpct, altfreq, variants, None)
bamobj.close()
usage()
elif not ref and os.path.exists(arg) and os.path.exists('%s.fai' % arg):
if os.path.exists('%s.fai' % arg):
ref = arg
else:
print "Missing FAI index on %s" % arg
usage()
elif not regions:
regions = BedFile(region=arg)
else:
print "Unknown option or missing index: %s" % arg
usage()
except Exception, e:
print e
usage()
if not bam:
usage()
else:
bamobj = bam_open(bam)
if profile:
import cProfile
def func():
bam_basecall(bamobj, ref, min_qual, min_count, regions, mask, quiet, showgaps, showstrand, minorpct, altfreq, variants, TimedProfiler())
sys.stderr.write('Profiling...\n')
cProfile.run('func()', profile)
else:
bam_basecall(bamobj, ref, min_qual, min_count, regions, mask, quiet, showgaps, showstrand, minorpct, altfreq, variants, None)
bamobj.close()
Region should be: chr:start-end (start 1-based)
"""
sys.exit(1)
if __name__ == "__main__":
fname = None
ref = None
start = None
end = None
for arg in sys.argv[1:]:
if arg == '-h':
usage()
elif not fname:
if os.path.exists(arg):
fname = arg
else:
usage("%s doesn't exist!")
else:
chrom, se = arg.split(':')
start, end = [int(x) for x in se.split('-')]
start = start - 1
if not fname:
usage()
bamfile = bam_open(fname)
bam_junction_count(bamfile, ref, start, end)
bamfile.close()
def bam_stats(infiles, gtf_file=None, region=None, delim=None, tags=[], show_all=False, fillin_stats=True):
if gtf_file:
gtf = GTF(gtf_file)
else:
gtf = None
sys.stderr.write('Calculating Read stats...\n')
stats = [BamStats(bam_open(x), gtf, region, delim, tags, show_all=show_all) for x in infiles]
sys.stdout.write('\t')
for fname, stat in zip(infiles, stats):
sys.stdout.write('%s\t\t' % fname)
sys.stdout.write('\n')
sys.stdout.write('Reads:\t')
for stat in stats:
sys.stdout.write('%s\t\t' % stat.total)
sys.stdout.write('\n')
sys.stdout.write('Mapped:\t')
for stat in stats:
sys.stdout.write('%s\t\t' % stat.mapped)
sys.stdout.write('\n')
sys.stdout.write('Unmapped:\t')
for stat in stats:
sys.stdout.write('%s\t\t' % stat.unmapped)
sys.stdout.write('\n')
sys.stdout.write('\nFlag distribution\n')
validflags = set()
maxsize = 0
for flag in flag_descriptions:
for stat in stats:
if stat.flag_counts.counts[flag] > 0:
validflags.add(flag)
maxsize = max(maxsize, len(flag_descriptions[flag]))
for flag in sorted(validflags):
sys.stdout.write("[0x%03x] %-*s" % (flag, maxsize, flag_descriptions[flag]))
for stat in stats:
sys.stdout.write('\t%s\t%0.2f%%' % (stat.flag_counts.counts[flag], (float(stat.flag_counts.counts[flag]) * 100 / stat.total)))
sys.stdout.write('\n')
sys.stdout.write('\n')
if stats[0].tlen_counts:
sys.stdout.write('Template length:')
for stat in stats:
mean, stdev = counts_mean_stdev(stat.tlen_counts)
sys.stdout.write('\t%0.2f\t+/- %0.2f' % (mean, stdev))
sys.stdout.write('\n')
sys.stdout.write('\n')
stat_tags = {}
for tag in stats[0].tagbins:
stat_tags[tag] = []
for stat in stats:
stat_tags[tag].append(stat.tagbins[tag])
for tag in stat_tags:
asc = stats[0].tagbins[tag].asc
sys.stdout.write("Ave %s:" % tag)
for i, tagbin in enumerate(stat_tags[tag]):
sys.stdout.write('\t%s' % tagbin.mean)
if i != len(stats):
sys.stdout.write('\t')
sys.stdout.write('\n')
sys.stdout.write("Max %s:" % tag)
for i, tagbin in enumerate(stat_tags[tag]):
sys.stdout.write('\t%s' % tagbin.max)
if i != len(stats):
sys.stdout.write('\t')
sys.stdout.write('\n')
sys.stdout.write('%s distribution:\n' % tag)
gens = []
gen_vals = []
last_pcts = []
for stat in stats:
gens.append(stat.distribution_gen(tag))
gen_vals.append(None)
last_pcts.append(0.0)
good = True
last = None
while good:
good = False
for i, stat in enumerate(stats):
if not gen_vals[i]:
try:
gen_vals[i] = gens[i].next()
except StopIteration:
pass
vals = [tup[0] for tup in gen_vals if tup]
if not vals:
continue
if asc:
minval = min(vals)
else:
minval = max(vals)
if last and type(last) == int and fillin_stats:
if asc:
last += 1
# fill in missing values
while last < minval:
sys.stdout.write('%s' % last)
for i, stat in enumerate(stats):
sys.stdout.write('\t0\t%s' % last_pcts[i])
sys.stdout.write('\n')
last += 1
else:
last -= 1
# fill in missing values
while last > minval:
sys.stdout.write('%s' % last)
for i, stat in enumerate(stats):
sys.stdout.write('\t0\t%s' % last_pcts[i])
sys.stdout.write('\n')
last -= 1
last = minval
sys.stdout.write(str(minval))
for i, tup in enumerate(gen_vals):
if tup and tup[0] == minval:
sys.stdout.write('\t%s\t%s' % (tup[1], tup[2]))
last_pcts[i] = tup[2]
gen_vals[i] = None
good = True
else:
sys.stdout.write('\t0\t%s' % (last_pcts[i]))
sys.stdout.write('\n')
sys.stdout.write('\n')
sys.stdout.write('Reference counts')
for stat in stats:
sys.stdout.write('\tcount\t')
sys.stdout.write('\n')
for k in sorted([x for x in stats[0].refs]):
sys.stdout.write('%s' % k)
for stat in stats:
sys.stdout.write('\t%s\t' % stat.refs[k])
sys.stdout.write('\n')
if gtf_file:
sys.stdout.write('Mapping regions')
for stat in stats:
sys.stdout.write('\tcount\tCPM')
sys.stdout.write('\n')
sorted_keys = [x for x in stats[0].regiontagger.counts]
sorted_keys.sort()
for k in sorted_keys:
sys.stdout.write('%s' % k)
for stat in stats:
sys.stdout.write('\t%s\t%s' % (stat.regiontagger.counts[k], float(stat.regiontagger.counts[k]) / stat.mapped / 1000000))
sys.stdout.write('\n')
usage()
elif not bamfile:
if not os.path.exists(arg):
usage("Missing or non-existant bamfile: %s" % arg)
if not os.path.exists("%s.bai" % arg):
usage("Missing bam index (bai) file: %s" % arg)
bamfile = arg
if not model or not model_arg:
usage("Missing model! Must include one of: %s" % ", ".join(count.models))
elif not bamfile:
usage("Missing BAM file!")
modelobj = count.models[model](model_arg)
bam = bam_open(bamfile)
modelobj.count(
bam,
stranded,
coverage,
uniq_only,
fpkm,
norm,
multiple,
whitelist,
blacklist,
rev_read2=rev_read2,
start_only=startonly,
)
bam.close()
last = arg
elif arg in ['-norm', '-multiple', '-whitelist', '-blacklist', '-library']:
last = arg
elif arg == '-startonly':
startonly = True
elif arg == '-coverage':
coverage = True
elif arg == '-fpkm':
fpkm = True
elif arg == '-uniq':
uniq_only = True
elif arg == '-h':
usage()
elif not bamfile:
if not os.path.exists(arg):
usage('Missing or non-existant bamfile: %s' % arg)
if not os.path.exists('%s.bai' % arg):
usage('Missing bam index (bai) file: %s' % arg)
bamfile = arg
if not model or not model_arg:
usage('Missing model! Must include one of: %s' % ', '.join(count.models))
elif not bamfile:
usage('Missing BAM file!')
modelobj = count.models[model](model_arg)
bam = bam_open(bamfile)
modelobj.count(bam, library_type, coverage, uniq_only, fpkm, norm, multiple, whitelist, blacklist, start_only=startonly)
bam.close()