if feature_coords_fn:
feature_coords = load(open(feature_coords_fn))
print >> stderr, "Note: using FASTA file at %s" % fastafn
f = open(gtffn)
junctions = set()
for row in f: # for each row in the input GTF
# we need to get rid of first and last exons
if row.strip().split('\t')[2] in ['start_codon', 'stop_codon', 'CDS']:
continue
if feature_coords_fn: # if we have the resource to check whether this is a terminal exon or not
# print >> stderr,row.strip()
strand = row.strip().split('\t')[6]
feature, blank = parse_lines([row.strip()], strand, get_transcripts)
# print >> stderr,feature
is_terminal_feature, up_exon, down_exon = terminal_exon(
list(feature)[0], feature_coords)
if is_terminal_feature: # if this is terminal ignore it
continue
# now we know that this is not a terminal exon so there must be neighbouring exons
up_exon_gtf = region_to_GTF(up_exon, feature_coords[up_exon],
get_transcripts)
down_exon_gtf = region_to_GTF(down_exon, feature_coords[down_exon],
get_transcripts)
if get_introns:
up_exon_five, up_exon_three = GTFrow_to_5p3pcoords(
up_exon_gtf, offset, use_chromnames)
five, three = GTFrow_to_5p3pcoords(row, offset, use_chromnames)
#!/home/paulk/software/bin/python
from __future__ import division
from sys import argv,exit,stderr
from subprocess import Popen,PIPE
from key_functions import parse_lines
import pysam
f = open(argv[1])
t = pysam.Tabixfile("resources/Homo_sapiens.GRCh37.66.gtf.gz")
for row in f:
if row[0] == 'u': continue
l = row.strip().split('\t')
regions = [l[0],l[1],l[5],l[6]]
the_exons = ""
for region in regions:
result = t.fetch(region[3:-2])
strand = region[-1]
exons,no_lines = parse_lines(result,strand,get_transcripts=True)
the_exons += ",".join(exons) + "\t"
print row.strip()+"\t"+the_exons
f.close()
get_introns = args.introns
if feature_coords_fn:
feature_coords = load(open(feature_coords_fn))
print >> stderr,"Note: using FASTA file at %s" % fastafn
f = open(gtffn)
junctions = set()
for row in f: # for each row in the input GTF
# we need to get rid of first and last exons
if row.strip().split('\t')[2] in ['start_codon','stop_codon','CDS']: continue
if feature_coords_fn: # if we have the resource to check whether this is a terminal exon or not
# print >> stderr,row.strip()
strand = row.strip().split('\t')[6]
feature,blank = parse_lines([row.strip()],strand,get_transcripts)
# print >> stderr,feature
is_terminal_feature,up_exon,down_exon = terminal_exon(list(feature)[0],feature_coords)
if is_terminal_feature: # if this is terminal ignore it
continue
# now we know that this is not a terminal exon so there must be neighbouring exons
up_exon_gtf = region_to_GTF(up_exon,feature_coords[up_exon],get_transcripts)
down_exon_gtf = region_to_GTF(down_exon,feature_coords[down_exon],get_transcripts)
if get_introns:
up_exon_five,up_exon_three = GTFrow_to_5p3pcoords(up_exon_gtf,offset,use_chromnames)
five,three = GTFrow_to_5p3pcoords(row,offset,use_chromnames)
down_exon_five,down_exon_three = GTFrow_to_5p3pcoords(down_exon_gtf,offset,use_chromnames)
junctions.add((up_exon_five,three,five,down_exon_three))
else:
five,three = GTFrow_to_5p3pcoords(row,offset,use_chromnames)
if chrom == '---': continue
st,sp = st_sp.split('-')
if not chrom_names: # if tabix needs 'chr' removed
chrom = chrom[3:]
else: # if tabix can work with 'chr'
pass
if chrom == "chrM": chrom = "chrMT"
if chrom == "M": chrom = "MT"
try:
lines = tabix_file_ptr.fetch(region="%s:%s-%s" % (chrom,st,sp))
except ValueError:
ps_missing.append((int(l[0]),ps_detail[int(l[colno])]))
# print l[0]
continue
gene_exons,no_lines = parse_lines(lines,sd,get_transcripts)
if gene_exons == []:
ps_missing.append((int(l[0]),ps_detail[int(l[colno])]))
#print l[0]
continue
if report:
print >> h,l[colno]+"\t"+",".join(gene_exons)
# for ge in gene_exons:
# print >> h,l[colno]+"\t"+ge
# """
# """
#if len(gene_exons) > 0: print >> g,"\t".join([",".join(gene_exons)]+l[colno+1:])
if len(gene_exons) > 0:
ge = random.choice(list(gene_exons))