def main(inFileName, geneList=[]):
dataH = {}
# nameL = ('Mutation GRCh37 genome position', 'Mutation GRCh37 strand','Gene name','ID_sample','ID_tumour','Primary site', \
# 'Site subtype','Primary histology','Histology subtype','Genome-wide screen','Mutation ID','Mutation CDS','Mutation AA', \
# 'Mutation Description','Mutation zygosity','Mutation somatic status','Pubmed_PMID','Sample source','Tumor origin','Comments')
nameL = ('Gene name', 'Mutation CDS', 'Mutation AA',
'Mutation Description', 'Mutation GRCh37 genome position',
'Mutation GRCh37 strand', 'Mutation somatic status')
inFile = open(inFileName)
headerL = inFile.readline()[:-1].split('\t')
idxH = dict([(x, headerL.index(x)) for x in nameL])
for line in inFile:
valueL = line[:-1].split('\t')
geneN = valueL[idxH['Gene name']]
if '_ENST' in geneN:
geneN = geneN.split('_ENST')[0]
if len(geneList) > 0 and geneN not in geneList:
continue
coord = valueL[idxH['Mutation GRCh37 genome position']]
if not coord:
continue
somatic = valueL[idxH['Mutation somatic status']]
if not 'somatic' in somatic:
continue
(chrNum, chrSta, chrEnd) = re.search('([^:-]+):([^:-]+)-([^:-]+)',
coord).groups()
cds = valueL[idxH['Mutation CDS']]
aa = valueL[idxH['Mutation AA']]
desc = valueL[idxH['Mutation Description']]
strand = valueL[idxH['Mutation GRCh37 strand']]
rm = re.match('c\.[\+\-_0-9]+([atgcATGC]*)(>|ins|del)([atgcATGC]*)',
cds)
if rm:
(ref, vtype, alt) = rm.groups()
else:
ref, alt = '', ''
if strand == '-':
ref = mybasic.rc(ref)
alt = mybasic.rc(alt)
chr = chrNum
if chr == '23':
chr = 'X'
chrNum = 'X'
elif chr == '24':
chr = 'Y'
chrNum = 'Y'
elif chr == '25':
chr = 'M'
chrNum = 'M'
# if vtype == 'del':
# rm = re.search('([ACGT]+)', alt.upper())
# ## if deleted bases are specified
# if alt != '' and rm:
# ## check if deleted bases are the same as reference sequences at the location
# new_ref = os.popen('samtools faidx /data1/Sequence/ucsc_hg19/hg19.fa chr%s:%s-%s' % (chr,chrSta,chrEnd)).readlines()[1:]
# new_ref = "".join(map(lambda x: x.rstrip().upper(), new_ref))
# if new_ref == alt.upper():
# chrSta = str(int(chrSta) - 1)
# ref = os.popen('samtools faidx /data1/Sequence/ucsc_hg19/hg19.fa chr%s:%s-%s' % (chr,chrSta,chrEnd)).readlines()[1:]
# ref = "".join(map(lambda x: x.rstrip().upper(), ref))
# alt = ref[0]
key = (chrNum, chrSta, chrEnd, strand, ref, alt)
if key in dataH:
mybasic.pushHash(dataH[key], 'geneN', geneN)
mybasic.pushHash(dataH[key], 'cds', cds)
mybasic.pushHash(dataH[key], 'aa', aa)
mybasic.pushHash(dataH[key], 'desc', desc)
else:
dataH[key] = {
'geneN': set([geneN]),
'cds': set([cds]),
'aa': set([aa]),
'desc': set([desc])
}
for ((chrNum, chrSta, chrEnd, strand, ref, alt),
infoH) in dataH.iteritems():
sys.stdout.write('chr%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n' % (chrNum,chrSta,chrEnd,strand, ref,alt,\
','.join(filter(lambda x: not x.startswith('ENSG'), list(infoH['geneN']))), ','.join(infoH['cds']), ','.join(infoH['aa']), ','.join(infoH['desc'])))
def parse_info(info, ref, indexH):
itemL = info.split(',')
resH = {}
for item in itemL:
arr = item.split('|')
if arr[indexH['Feature_type']] == 'RegulatoryFeature':
gene = '-'
if gene not in resH:
resH[gene] = {'ch_type':''}
mybasic.pushHash(resH[gene], 'ch_type', arr[indexH['Consequence']])
mybasic.pushHash(resH[gene], 'strand', '*')
elif arr[indexH['Feature_type']] == 'MotifFeature' and (arr[indexH['Consequence']] == 'TF_binding_site_variant' or 'TFBS_' in arr[indexH['Consequence']]):
gene = '-'
if gene not in resH:
resH[gene] = {'ch_type':''}
mybasic.pushHash(resH[gene], 'ch_type', arr[indexH['Consequence']])
mybasic.pushHash(resH[gene], 'strand', '*')
elif arr[indexH['Feature_type']] == '' and (arr[indexH['Consequence']] == 'intergenic_variant'):
gene = '-'
if gene not in resH:
resH[gene] = {'ch_type':''}
mybasic.pushHash(resH[gene], 'ch_type', arr[indexH['Consequence']])
mybasic.pushHash(resH[gene], 'strand', '*')
elif arr[indexH['Feature_type']] == 'Transcript':
csq = arr[indexH['Consequence']]
# if ('non_coding_exon_variant' in csq or 'nc_transcript_variant' in csq) and 'splice_' not in csq and 'miRNA' not in csq:
# continue ## non-coding scripts other than miRNA
if ('upstream_gene_variant' in csq or 'downstream_gene_variant' in csq) and 'splice_' not in csq:
gene = '-'
if gene not in resH:
resH[gene] = {'ch_type':''}
## treat up-, down-stream gene variants as intergenic
mybasic.pushHash(resH[gene], 'strand', '*')
mybasic.pushHash(resH[gene], 'ch_type', 'intergenic_variant')
continue
# if 'intron_variant' in csq and 'splice_' not in csq:
# continue ## intron
csq = shorten_csq(csq)
gene = arr[indexH['SYMBOL']]
if gene not in resH:
resH[gene] = {'ch_type':''}
ch_dna = arr[indexH['HGVSc']]
if len(ch_dna.split(':')) > 1:
ch_dna = ch_dna.split(':')[1]
prot_pos = arr[indexH['Protein_position']]
aa = arr[indexH['Amino_acids']]
if arr[indexH['CANONICAL']] != 'YES':
aa = ''
if len(aa) < 1:
# if len(prot_pos) > 0:
# print info
# print 'AA:%s' % item
# raise Exception
ch_aa = ''
elif '/' not in aa:
# elif len(aa) < 2:
ch_aa = 'p.%s%s%s' % (aa, prot_pos, aa)
else:
(aa1, aa2) = aa.split('/')
ch_aa = 'p.%s%s%s' % (aa1, prot_pos, aa2)
codon = arr[indexH['Codons']]
if len(codon) > 0:
(nt1,nt2) = re.match('[nacgt]*([-ACGT]*)[nacgt]*/[nacgt]*([-ACGT]*)[nacgt]*', codon).group(1,2)
if ref != nt1:
strand = '-'
else:
strand = '+'
if 'strand' in resH[gene]:
try:
resH[gene]['strand'].remove('*')
except:
pass
mybasic.pushHash(resH[gene], 'strand', strand)
else:
if 'strand' not in resH[gene]:
mybasic.pushHash(resH[gene], 'strand', '*')
if len(csq) > 0:
if len(arr[indexH['CANONICAL']]) > 0:
mybasic.pushHash(resH[gene], 'ch_type_C', csq)
mybasic.pushHash(resH[gene], 'ch_type', csq)
if len(ch_aa) > 0 and 'nc_transcript_' not in csq and 'non_coding_' not in csq:
if len(arr[indexH['CANONICAL']]) > 0:
mybasic.pushHash(resH[gene], 'ch_aa_C', ch_aa)
mybasic.pushHash(resH[gene],'ch_aa', ch_aa)
if len(ch_dna) > 0 and 'nc_transcript_' not in csq and 'non_coding_' not in csq:
if len(arr[indexH['CANONICAL']]) > 0:
mybasic.pushHash(resH[gene], 'ch_dna_C', ch_dna)
mybasic.pushHash(resH[gene], 'ch_dna', ch_dna)
return resH
def main(inFileName,geneList=[]):
dataH = {}
# nameL = ('Mutation GRCh37 genome position', 'Mutation GRCh37 strand','Gene name','ID_sample','ID_tumour','Primary site', \
# 'Site subtype','Primary histology','Histology subtype','Genome-wide screen','Mutation ID','Mutation CDS','Mutation AA', \
# 'Mutation Description','Mutation zygosity','Mutation somatic status','Pubmed_PMID','Sample source','Tumor origin','Comments')
nameL = ('Gene name','Mutation CDS','Mutation AA','Mutation Description','Mutation GRCh37 genome position','Mutation GRCh37 strand','Mutation somatic status')
inFile = open(inFileName)
headerL = inFile.readline()[:-1].split('\t')
idxH = dict([(x, headerL.index(x)) for x in nameL])
for line in inFile:
valueL = line[:-1].split('\t')
geneN = valueL[idxH['Gene name']]
if '_ENST' in geneN:
geneN = geneN.split('_ENST')[0]
if len(geneList)>0 and geneN not in geneList:
continue
coord = valueL[idxH['Mutation GRCh37 genome position']]
if not coord:
continue
somatic = valueL[idxH['Mutation somatic status']]
if not 'somatic' in somatic:
continue
(chrNum,chrSta,chrEnd) = re.search('([^:-]+):([^:-]+)-([^:-]+)', coord).groups()
cds = valueL[idxH['Mutation CDS']]
aa = valueL[idxH['Mutation AA']]
desc = valueL[idxH['Mutation Description']]
strand = valueL[idxH['Mutation GRCh37 strand']]
rm = re.match('c\.[\+\-_0-9]+([atgcATGC]*)(>|ins|del)([atgcATGC]*)',cds)
if rm:
(ref,vtype,alt) = rm.groups()
else:
ref,alt = '',''
if strand == '-':
ref = mybasic.rc(ref)
alt = mybasic.rc(alt)
chr = chrNum
if chr == '23':
chr = 'X'
chrNum = 'X'
elif chr == '24':
chr = 'Y'
chrNum = 'Y'
elif chr == '25':
chr = 'M'
chrNum = 'M'
# if vtype == 'del':
# rm = re.search('([ACGT]+)', alt.upper())
# ## if deleted bases are specified
# if alt != '' and rm:
# ## check if deleted bases are the same as reference sequences at the location
# new_ref = os.popen('samtools faidx /data1/Sequence/ucsc_hg19/hg19.fa chr%s:%s-%s' % (chr,chrSta,chrEnd)).readlines()[1:]
# new_ref = "".join(map(lambda x: x.rstrip().upper(), new_ref))
# if new_ref == alt.upper():
# chrSta = str(int(chrSta) - 1)
# ref = os.popen('samtools faidx /data1/Sequence/ucsc_hg19/hg19.fa chr%s:%s-%s' % (chr,chrSta,chrEnd)).readlines()[1:]
# ref = "".join(map(lambda x: x.rstrip().upper(), ref))
# alt = ref[0]
key = (chrNum,chrSta,chrEnd,strand,ref,alt)
if key in dataH:
mybasic.pushHash(dataH[key],'geneN',geneN)
mybasic.pushHash(dataH[key],'cds',cds)
mybasic.pushHash(dataH[key],'aa',aa)
mybasic.pushHash(dataH[key],'desc',desc)
else:
dataH[key] = {'geneN':set([geneN]), 'cds':set([cds]), 'aa':set([aa]), 'desc':set([desc])}
for ((chrNum,chrSta,chrEnd,strand,ref,alt),infoH) in dataH.iteritems():
sys.stdout.write('chr%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n' % (chrNum,chrSta,chrEnd,strand, ref,alt,\
','.join(filter(lambda x: not x.startswith('ENSG'), list(infoH['geneN']))), ','.join(infoH['cds']), ','.join(infoH['aa']), ','.join(infoH['desc'])))
def parse_info_old(info, indexH):
itemL = info.split(',')
resH = {}
for item in itemL:
arr = item.split('|')
if arr[indexH['Consequence']] == 'regulatory_region_variant': ## exclude variants in regulatory region (for now)
continue
key = arr[indexH['Gene']]
tid = arr[indexH['Feature']]
if key in resH.keys():
if tid != '':
if tid in resH[key]['tid']:
##something wrong!!
print info
sys.exit(1)
else:
tnum = len(resH[key]['tid']) + 1
resH[key]['tid'][tid] = str(tnum)+":"+tid
else:
tnum = 1
##
for k in indexH.keys():
if arr[indexH[k]] != '':
if k in excField:
continue
elif k in sglField:
mybasic.pushHash(resH[key], k, arr[indexH[k]])
else:
if tid == '':
mybasic.pushHash(resH[key], k, arr[indexH[k]])
else:
if k == 'HGVSc' or k == 'HGVSp':
val = arr[indexH[k]].split(':')[1]
mybasic.pushHash(resH[key], k, str(tnum)+"="+val)
else:
mybasic.pushHash(resH[key], k, str(tnum)+"="+arr[indexH[k]])
else:
resH[key] = {}
resH[key]['tid'] = {}
if tid != '':
tnum = len(resH[key]['tid'])+1
resH[key]['tid'][tid] = str(tnum)+":"+tid
for k in indexH.keys():
if arr[indexH[k]] != '':
if k in excField:
continue
elif k in sglField:
mybasic.pushHash(resH[key], k, arr[indexH[k]])
else:
if tid == '':
mybasic.pushHash(resH[key], k, arr[indexH[k]])
else:
if k == 'HGVSc' or k == 'HGVSp':
val = arr[indexH[k]].split(':')[1]
mybasic.pushHash(resH[key], k, str(tnum)+"="+val)
else:
mybasic.pushHash(resH[key], k, str(tnum)+"="+arr[indexH[k]])
return resH