# save the data file
# expfile
"""
S1 S2 .. Sn
G1 ...
G2 ...
"""
expreader = TsvReader(expfile)
expdata = [r for r in expreader if r[0] in genes or r[0] in tfs]
expreader.close()
datawriter = TsvWriter(outdata)
for i, cname in enumerate(expreader.cnames):
if i == 0:
# genes + tfs
datawriter.cnames = [r[0] for r in expdata]
datawriter.writeHead()
else:
datawriter.write([cname] + [r[i] for r in expdata])
datawriter.close()
del expdata
genes = [g for g in genes if g in datawriter.cnames]
tfs = [g for g in tfs if g in datawriter.cnames]
genetfs = {g: [tf for tf in gtfs if tf in tfs] for g, gtfs in genetfs.items() if g in genes}
# save the group file
# mutfile
"""
S1 S2 .. Sn
M1 ... (0/1/2/NA)
M2 ...
indata1 = TsvReader(infile1, **inopts1) indata2 = TsvReader(infile2, **inopts2) cnames1 = indata1.meta if not rnames1 else indata1.meta[1:] cnames2 = indata2.meta if not rnames2 else indata2.meta[1:] paired = list(set(cnames1) & set(cnames2)) cnames1 = cnames2 = paired if rnames1: cnames1 = [indata1.meta[0]] + cnames1 if rnames2: cnames2 = [indata2.meta[0]] + cnames2 cindex1 = [indata1.meta.index(c) for c in cnames1] cindex2 = [indata2.meta.index(c) for c in cnames2] outdata1 = TsvWriter(outfile1) outdata2 = TsvWriter(outfile2) outdata1.meta = cnames1 outdata2.meta = cnames2 outdata1.writeHead() outdata2.writeHead() for r1 in indata1: outdata1.write(r1[i] for i in cindex1) outdata1.close() for r2 in indata2: outdata2.write(r2[i] for i in cindex2) outdata2.close()
cmdparams = []
params.thresh = pval
params.verbosity = 4
for motif, name in motifs.items():
params.oc = path.join(outdir, name + '.' + re.sub(r'[^\w_]', '', motif))
params.motif = motif
params[""] = [tfmotifs, sfile]
cmdparams.append((meme, cmdargs(params, dash = '--', equal = ' ')))
Parallel(nthread, raiseExc = True).run('{} {}', cmdparams)
writer = TsvWriter(outfile)
writer.cnames = [
"CHR", "START", "END", "NAME", "SCORE", "STRAND", "MOTIF", "SEQ", "STARTONSEQ",
"STOPONSEQ", "RAWSCORE", "PVAL", "QVAL", "MATCHEDSEQ", "UCSCLINK"
]
writer.writeHead(callback = lambda cnames: "#" + "\t".join(cnames))
def rowfactory(r):
r.PVAL = float(r['p-value'])
if r.PVAL >= pval:
return None
r.RAWSCORE = r.score
try:
r.SCORE = int(float(r.score) * 10)
except TypeError:
r.SCORE = 0
r.STRAND = r.strand
r.MOTIF = r.motif_id
# split motif_alt_id
# GENE or GENE::chr1:111-222 or ::chr1:111-222 or chr1:111-222
r.SEQ = r.sequence_name
outfile = {{ o.outfile | quote}}
outdir = {{ o.outdir | quote}}
params = {{ args.params | repr}}
idxfile = {{ args.idxfile | quote}}
kallisto = {{ args.kallisto | quote}}
nthread = {{ args.nthread | repr}}
shell.TOOLS.kallisto = kallisto
params.i = idxfile
params.o = outdir
params.t = nthread
params._ = [fq1, fq2]
kallisto = shell.Shell(subcmd = True).kallisto
kallisto.quant(**params).run()
imfile = path.join(outdir, 'abundance.tsv')
reader = TsvReader(imfile)
writer = TsvWriter(outfile)
writer.cnames = ['target_id', 'est_counts']
writer.writeHead()
for r in reader:
r.target_id = r.target_id.split('::')[0]
try:
r.est_counts = int(round(float(r.est_counts)))
except TypeError:
r.est_counts = 0
writer.write(r)
writer.close()
for _ in range(dist):
writer.write(next(reader))
writer.close()
para = Parallel(nthread, raiseExc = True)
para.run(getAlleleCount, [
(tumbam, path.join(
thdir, '{bname}.thread{i}.snp'.format(bname = asbname, i = i)
), path.join(
thdir, '{tumbn}.thread{i}.bamrc'.format(tumbn = path.basename(tumbam), i = i)
)) for i in range(nthread)
])
# merge to tumsnp
writer = TsvWriter(tumsnp)
writer.cnames = ['Chrm', 'pos', 'A', 'C', 'G', 'T', 'Total', 'refCount', 'mutCount']
writer.writeHead(lambda cn: "#" + "\t".join(cn))
for i in range(nthread):
subrc = path.join(
thdir, '{tumbn}.thread{i}.bamrc'.format(tumbn = path.basename(tumbam), i = i)
)
reader = TsvReader(subrc, cnames = False)
for r in reader:
writer.write(r.values())
reader.close()
writer.close()
# normal
para.run(getAlleleCount, [
(normbam, path.join(
thdir, '{bname}.thread{i}.snp'.format(bname = asbname, i = i)
), path.join(