def get_lengths(args, tdir):
# get our gene lengths
lengths = {}
of = open(tdir + '/ref.bed', 'w')
cmd = 'bedtools sort -i - | bedtools merge -i - -c 4 -o collapse'
p = subprocess.Popen(cmd, shell=True, stdin=subprocess.PIPE, stdout=of)
with open(args.transcriptome_reference) as inf:
for line in inf:
if re.match('^#', line): continue
e = genepred_line_to_entry(line)
for i in range(0, len(e['exonStarts'])):
#dont' consider exons that are too long.
if e['exonEnds'][i] - e['exonStarts'][
i] > args.max_exon_length and args.max_exon_length > 0:
continue
p.stdin.write(e['chrom'] + "\t" + str(e['exonStarts'][i]) +
"\t" + str(e['exonEnds'][i]) + "\t" +
e['gene_name'] + "\n")
p.communicate()
of.close()
with open(tdir + '/ref.bed') as inf:
for line in inf:
f = line.rstrip().split("\t")
elen = int(f[2]) - int(f[1])
genes = f[3].split(',')
for gene in genes:
if gene not in lengths:
lengths[gene] = 0
lengths[gene] += elen
return lengths
def transcriptome_to_exons(fname, tdir):
of1 = open(tdir + '/all_exons.bed', 'w')
of2 = open(tdir + '/all_loci.bed', 'w')
bounds = {}
with open(fname) as inf:
for line in inf:
if re.match('^#', line): continue
e = genepred_line_to_entry(line)
for i in range(0, len(e['exonStarts'])):
if e['chrom'] not in bounds:
bounds[e['chrom']] = [100000000000, 0]
if e['exonStarts'][i] < bounds[e['chrom']][0]:
bounds[e['chrom']][0] = e['exonStarts'][i]
if e['exonEnds'][i] > bounds[e['chrom']][1]:
bounds[e['chrom']][1] = e['exonEnds'][i]
of1.write(e['chrom'] + "\t" + str(e['exonStarts'][i]) + "\t" +
str(e['exonEnds'][i]) + "\n")
of2.write(e['chrom'] + "\t" + str(e['txStart']) + "\t" +
str(e['txEnd']) + "\n")
of1.close()
of2.close()
# Get the compressed exons
cmd = "bedtools sort -i " + tdir + '/all_exons.bed > ' + tdir + '/all_exons.sorted.bed'
subprocess.call(cmd, shell=True)
cmd = "bedtools merge -i " + tdir + '/all_exons.sorted.bed > ' + tdir + '/merged_exons.bed'
subprocess.call(cmd, shell=True)
cmd = "bedtools sort -i " + tdir + '/all_loci.bed > ' + tdir + '/all_loci.sorted.bed'
subprocess.call(cmd, shell=True)
cmd = "bedtools merge -i " + tdir + '/all_loci.sorted.bed > ' + tdir + '/merged_loci.bed'
subprocess.call(cmd, shell=True)
return bounds
def transcriptome_to_exons(fname,tdir):
of1 = open(tdir+'/all_exons.bed','w')
of2 = open(tdir+'/all_loci.bed','w')
bounds = {}
with open(fname) as inf:
for line in inf:
if re.match('^#',line): continue
e = genepred_line_to_entry(line)
for i in range(0,len(e['exonStarts'])):
if e['chrom'] not in bounds:
bounds[e['chrom']] = [100000000000,0]
if e['exonStarts'][i] < bounds[e['chrom']][0]:
bounds[e['chrom']][0] = e['exonStarts'][i]
if e['exonEnds'][i] > bounds[e['chrom']][1]:
bounds[e['chrom']][1] = e['exonEnds'][i]
of1.write(e['chrom']+"\t"+str(e['exonStarts'][i])+"\t"+str(e['exonEnds'][i])+"\n")
of2.write(e['chrom']+"\t"+str(e['txStart'])+"\t"+str(e['txEnd'])+"\n")
of1.close()
of2.close()
# Get the compressed exons
cmd = "bedtools sort -i "+tdir+'/all_exons.bed > '+tdir+'/all_exons.sorted.bed'
subprocess.call(cmd,shell=True)
cmd = "bedtools merge -i "+tdir+'/all_exons.sorted.bed > '+tdir+'/merged_exons.bed'
subprocess.call(cmd,shell=True)
cmd = "bedtools sort -i "+tdir+'/all_loci.bed > '+tdir+'/all_loci.sorted.bed'
subprocess.call(cmd,shell=True)
cmd = "bedtools merge -i "+tdir+'/all_loci.sorted.bed > '+tdir+'/merged_loci.bed'
subprocess.call(cmd,shell=True)
return bounds
def main():
parser = argparse.ArgumentParser()
parser.add_argument('input',help='FILENAME input genepred, use - for STDIN')
args = parser.parse_args()
inf = sys.stdin
if args.input != '-':
inf = open(args.input)
for line in inf:
if re.match('^#',line): continue
e = genepred_line_to_entry(line)
for i in range(0,len(e['exonStarts'])):
print e['chrom']+"\t"+str(e['exonStarts'][i])+"\t"+str(e['exonEnds'][i])+"\t"+e['gene_name']+"\t"+e['name']+"\t"+str(i)
inf.close()
def main():
parser = argparse.ArgumentParser()
parser.add_argument('input',
help='FILENAME input genepred, use - for STDIN')
args = parser.parse_args()
inf = sys.stdin
if args.input != '-':
inf = open(args.input)
for line in inf:
if re.match('^#', line): continue
e = genepred_line_to_entry(line)
for i in range(0, len(e['exonStarts'])):
print e['chrom'] + "\t" + str(e['exonStarts'][i]) + "\t" + str(
e['exonEnds']
[i]) + "\t" + e['gene_name'] + "\t" + e['name'] + "\t" + str(i)
inf.close()
def get_lengths(args, tdir):
# get our gene lengths
lengths = {}
of = open(tdir + "/ref.bed", "w")
cmd = "bedtools sort -i - | bedtools merge -i - -c 4 -o collapse"
p = subprocess.Popen(cmd, shell=True, stdin=subprocess.PIPE, stdout=of)
with open(args.transcriptome_reference) as inf:
for line in inf:
if re.match("^#", line):
continue
e = genepred_line_to_entry(line)
for i in range(0, len(e["exonStarts"])):
# dont' consider exons that are too long.
if e["exonEnds"][i] - e["exonStarts"][i] > args.max_exon_length and args.max_exon_length > 0:
continue
p.stdin.write(
e["chrom"]
+ "\t"
+ str(e["exonStarts"][i])
+ "\t"
+ str(e["exonEnds"][i])
+ "\t"
+ e["gene_name"]
+ "\n"
)
p.communicate()
of.close()
with open(tdir + "/ref.bed") as inf:
for line in inf:
f = line.rstrip().split("\t")
elen = int(f[2]) - int(f[1])
genes = f[3].split(",")
for gene in genes:
if gene not in lengths:
lengths[gene] = 0
lengths[gene] += elen
return lengths