#!/usr/bin/env python2.7
# -*- coding: utf-8 -*
'''
Author: Mei Hou
Get the transcript annotation from TangX's V4expCaculate.combined.gtf
Include: 'transcript_id', 'gene_id', 'gene_biotype', 'transcript_type', 'source', 'gene_name', 'transcript_name'
Note: if a annotation type is not existed for a transcript, it is '-'
'''
import sys
import modify_gtf as gtf
curTransId = ''
tAnnoType = ('transcript_id', 'gene_id', 'gene_biotype', 'transcript_type',
'source', 'gene_name', 'transcript_name')
print '\t'.join(tAnnoType)
for line in sys.stdin.readlines():
li = line.rstrip('\n').split('\t')
# 只看exon的，以防其他UTR这些注释不全
if li[2] == 'exon':
dAnnos = gtf.processAnnotation(li[8])
dAnnos['source'] = li[1]
if curTransId != dAnnos['transcript_id']:
curTransId = dAnnos['transcript_id']
lOut = []
for at in tAnnoType:
lOut.append(dAnnos.setdefault(at, '-'))
print '\t'.join(lOut)
dClusterAnno = {}
for ca in clusterAnnoFile.readlines():
ca = ca.rstrip('\n').split('\t')
dClusterAnno[ca[4]] = [ca[6]] + ca[9:12]
# tidy the intersect file
print '\t'.join(['ClusterID', 'Chromosome', 'Strand', 'ClusterLength', 'OverlapLength', 'TranscriptID', 'BioType', 'Feature', 'GeneName', 'GeneID', 'ReadCout', 'ConversionLocationCount', 'ConversionEventCount', 'NonConversionEventCount'])
intersectFile = open(intersectFile)
for li in intersectFile.readlines():
li = li.rstrip('\n').split('\t')
clusterId = li[12]
feature = li[2]
# need to count by biotype
dTransBasicAnno = gtf.processAnnotation(li[8])
transId = dTransBasicAnno['transcript_id']
# check
if not transId in dTransAnno:
sys.stderr.write(transId + ' is not in the annotation file!\n')
sys.exit(1)
# check
if not clusterId in dClusterAnno:
sys.stderr.write(clusterId + ' is not in the annotation file!\n')
sys.exit(1)
transAnno = dTransAnno[transId]
biotype = transAnno[1]
clusterAnno = dClusterAnno[clusterId]
#!/usr/bin/env python2.7
# -*- coding: utf-8 -*
import sys
import modify_gtf as gtf
from collections import defaultdict
ddTrans = defaultdict(int)
curTransId = ''
ok = 1
for line in sys.stdin.readlines():
li = line.rstrip('\n').split('\t')
dAnnos = gtf.processAnnotation(li[8])
lineTransId = dAnnos['transcript_id']
if lineTransId != curTransId:
curTransId = lineTransId
ddTrans[curTransId] += 1
for k, v in ddTrans.items():
if v > 1:
sys.stderr.write(k + ' is repeated!\n')
ok = 0
if ok == 1:
print 'The file is well sorted by transcripts!'
# tidy the intersect file
print '\t'.join([
'ClusterID', 'Chromosome', 'Strand', 'ClusterLength', 'OverlapLength',
'TranscriptID', 'BioType', 'Feature', 'GeneName', 'GeneID', 'ReadCout',
'ConversionLocationCount', 'ConversionEventCount',
'NonConversionEventCount'
])
intersectFile = open(intersectFile)
for li in intersectFile.readlines():
li = li.rstrip('\n').split('\t')
clusterId = li[12]
feature = li[2]
# need to count by biotype
dTransBasicAnno = gtf.processAnnotation(li[8])
transId = dTransBasicAnno['transcript_id']
# check
if not transId in dTransAnno:
sys.stderr.write(transId + ' is not in the annotation file!\n')
sys.exit(1)
# check
if not clusterId in dClusterAnno:
sys.stderr.write(clusterId + ' is not in the annotation file!\n')
sys.exit(1)
transAnno = dTransAnno[transId]
biotype = transAnno[1]
clusterAnno = dClusterAnno[clusterId]