def callCuffquant(bam, gtf, cpu=3):
fo = bam.split("/")[-2]
if os.path.exists(fo):
logger.info("%s has been generated to %s" % (bam, fo))
cmd = "cuffquant -p {cpu} -o {fout} {gtf} {bam}".format(cpu=cpu,
fout=fo,
gtf=gtf,
bam=bam)
callSys([cmd], logger)
def sam2bam(sam, bam):
"""
SAM to BAM file
"""
samview = "samtools view -S %s -b -o %s" % (sam, bam)
samsort = "samtools sort -@ 2 {bam} -T {pre} -o {bam}".format(
bam=bam, pre=bam.replace(".bam", ""))
samindex = "samtools index {bam} {bai}".format(bam=bam,
bai=bam.replace(
".bam", ".bai"))
rmsam = "rm %s" % (sam)
cmds = [samview, samsort, samindex, rmsam]
callSys(cmds, logger)
def bdg2bw(f):
"""
Converting .bdg file to .bw file through bedGraphToBigWig.
"""
n = f.split("/")[-1].replace(".bdg", "")
if os.path.isfile(n + ".bw"):
return
cmd1 = "bedSort {bdg} {sbdg}".format(bdg=f, sbdg=n + ".bdg2")
callSys([cmd1], logger)
#validation bdg
validateBdg(n + ".bdg2")
cmd2 = "bedGraphToBigWig {bdg} {chrom} {bw}".format(bdg=n + ".bdg2",
chrom=CHROM,
bw=n + ".bw")
callSys([cmd2, "rm %s.bdg2" % n], logger)
def bdg2washU(f):
"""
Converting .bdg file to washU track.
"""
n = f.split("/")[-1].replace(".bdg", "")
if os.path.isfile(n +
"_washU.txt.gz") and os.path.isfile(n +
"_washU.txt.gz.tbi"):
return
#sort the .bdg file
cmd1 = "bedSort {bdg} {sbdg}".format(bdg=f, sbdg=n + "_washU.txt")
#bgzip
cmd2 = "bgzip %s" % (n + "_washU.txt")
#tabix
cmd3 = "tabix -p bed %s" % (n + "_washU.txt.gz")
callSys([cmd1, cmd2, cmd3], logger)
def validateBdg(bdg):
"""
Validating .bdg files through chrom size.
"""
chrs = getChrSize()
nbdg = bdg + ".2"
with open(nbdg, "w") as f:
for line in open(bdg):
line = line.split("\n")[0].split("\t")
if len(line) < 4:
continue
if line[0] not in chrs:
continue
if int(line[1]) >= chrs[line[0]] or int(line[2]) > chrs[line[0]]:
continue
line = "\t".join(line) + "\n"
f.write(line)
cmd = "mv %s %s" % (nbdg, bdg)
callSys([cmd], logger)
def callCuffnorm(fs, gtf, fo, cpu=40):
cmd = "cuffnorm --compatible-hits-norm {gtf} {fs} -o {fo} -p {cpu}".format(
gtf=gtf, fs=" ".join(fs), fo=fo, cpu=cpu)
callSys([cmd], logger)