def get_indel_info(self):
'''处理indel vcf文件
处理indel的vcf文件时，为了和注释的pos,以及ref和alt对应，需要对vcf做处理
{
'chr_pos_ref/alt': [(case), (control)]
}
'''
with utils.safe_open(self.vcf, 'r') as fr:
for line in fr:
if line.startswith('##'):
pass
elif line.startswith('#'):
head = line.strip('')
head_index = utils.get_head_index(head)
continue
linelist = line.strip('').split('\t')
_chr = linelist[head_index['#chrom']]
pos = linelist[head_index['pos']]
ref = linelist[head_index['ref']]
alt = linelist[head_index['alt']]
pos, ref, alt = utils.modify_pos_ref_alt(pos, ref, alt)
if 'cancer' in head_index:
case = linelist[head_index['cancer']]
if 'normal' in head_index:
control = linelist[head_index['normal']]
def get_tran_relation(self):
'''
input:
self.transcript: gene transcript
output:
list: [gene=trans, gene=trans]
'''
tran_relation = []
with utils.safe_open(self.transript_database, 'r') as fr:
for line in fr:
if line.startswith('#'):
head_index = utils.get_head_index(line)
continue
linelist = line.strip('').split('\t')
gene = linelist[head_index['#gene']]
tran = linelist[head_index['transcript']]
tran_relation.append('{gene}={tran}'.format(**locals()))
return tran_relation
print(len(rules))
relation_to_body = {}
relation_to_body = add_relation_body(rules, relation_to_body)
print(len(relation_to_body))
# # Length 1 Rules
# In[126]:
dict_1 = utils.get_r_e1e2_dict(mapped_train_data)
# # Length 2 Rules
# In[127]:
index_head = utils.get_head_index(mapped_train_data)
dict_2 = utils.get_r1r2_e1e2_dict(mapped_train_data, index_head)
# ## Length 3 Rules
# In[128]:
## Gives entity in path for given relation and body
# In[129]:
re2_e1 = utils.get_re2_e1_dict(mapped_train_data)
e1r_e2 = utils.get_e1r_e2_dict(mapped_train_data)
# In[130]:
def start(self):
'''
程序运行主函数
'''
if self.vcf:
vcf_info = get_vcf_info.HandleVcf(self.vcf, self.vcftype).start()
with open(self.vep, 'r') as fr, open('test', 'w') as fw:
for line in fr:
if line.startswith('##'):
continue
elif line.startswith('#Uploaded_variation'):
head = line
head_index = utils.get_head_index(head)
continue
linelist = line.strip('').split('\t')
#获取需要的信息
row = headers.HEAD()
## 可以直接提取的信息
upload_variation = linelist[head_index[
'#Uploaded_variation'.lower()]] # chr5_1295229_-/A
location = linelist[
head_index['location']] # chr5:1295187-1295188
transcript = linelist[head_index['feature']] # NM_198253.3
function = linelist[
head_index['consequence']] # missense_variant
strand = linelist[head_index['strand']] # -1
gene = linelist[head_index['symbol']] # TERT
protein = linelist[head_index['ensp']] # NP_937983.2
sift = linelist[head_index['sift']] # tolerated(0.05)
polyphen = linelist[head_index['polyphen']]
exon_id = linelist[head_index['exon']] # 2/19 or -
chgvs = linelist[head_index['hgvsc']] # NM_198253.3:c.77C>T
phgvs = linelist[head_index['hgvsp']] # NP_937983.2:p.Thr26Met
tert = linelist[head_index['tert']] # 只有tert的启动子区域有
clinvar = linelist[head_index['clinvar_clnsig']]
rs = linelist[head_index['existing_variation']]
# 需要进行处理获取的信息
hgvsc = utils.simplify_hgvsc(chgvs)
hgvsp2 = utils.simplify_hgvsp(phgvs)
hgvsp = utils.get_oneletter_hgvsp(hgvsp2)
exon_id = utils.get_exon_id(exon_id)
_chr, start, end = utils.get_chr_start_end_from_location(
location)
ref, alt = utils.get_ref_alt_from_upload_variation(
upload_variation)
muttype = utils.get_muttype(ref, alt)
genotype = utils.get_genotype(ref, alt, strand)
flank = utils.get_flank_according_upload_variation(
upload_variation, self.hg19)
bl_muttype = utils.get_bl_muttype()
# 更新row
row.gene = gene
row.chgvs = hgvsc
row.phgvs = hgvsp
row.phgvs2 = hgvsp2
row.exon_id = exon_id
row.vep_function = function
row.sift = sift
row.polyphen2 = polyphen
row.chr = _chr
row.start = start
row.end = end
row.ref = ref
row.alt = alt
row.muttype = muttype
row.genotype = genotype
row.transcript = transcript
row.protein = protein
row.strand = strand
row.flank = flank
row.rs = rs
row.bl_muttype = bl_muttype
row.clinvar = clinvar
if self.vcf:
freq_tag = vcf_info[upload_variation]
print(freq_tag)
info = row.update_head(**freq_tag)
fw.write('\t'.join(info.keys()) + '\n')
fw.write('\t'.join(map(str, info.values())) + '\n')
if __name__ == '__main__':
import argparse
parser = argparse.ArgumentParser(description='test')
parser.add_argument('--infile')
parser.add_argument('--result')
parser.add_argument('--vep_function_yaml', help='func 配置文件')
args = vars(parser.parse_args())
infile = args['infile']
result = args['result']
with open(infile, 'r') as fr, open(result, 'w') as fw:
for line in fr:
linelist = line.strip('\n').split('\t')
if line.startswith('#'):
head_index = utils.get_head_index(linelist)
fw.write('{}\tvep_simple\tbgi_func\n'.format(
'\t'.join(linelist)))
continue
nm = linelist[head_index['feature']]
if not nm.startswith('NM'):
continue
gene = linelist[head_index['symbol']]
upload_variation = linelist[head_index[
'#Uploaded_variation'.lower()]]
vep_function = linelist[head_index['consequence']]
chgvs = linelist[head_index['hgvsc']]
phgvs = linelist[head_index['hgvsp']]
tert = linelist[head_index['tert']]
def get_snv_info(self):
'''处理snv vcf文件
读取INFO字段,获取需要的信息
vcf_info = {
chr_pos_ref/alt: [(case), (control)]
}
key的形式和vep注释结果中的upload_variation字段一致
'''
vcf_info = defaultdict(dict)
with utils.safe_open(self.vcf, 'r') as fr:
for line in fr:
if line.startswith('##'):
continue
elif line.startswith('#'):
head = line.strip('')
head_index = utils.get_head_index(head)
continue
linelist = line.strip('').split('\t')
_chr = linelist[head_index['#chrom']]
pos = linelist[head_index['pos']]
ref = linelist[head_index['ref']]
alt = linelist[head_index['alt']]
key = '{_chr}_{pos}_{ref}/{alt}'.format(**locals())
info = linelist[head_index['info']]
case = re.search(self.case_pattern, info)
control = re.search(self.control_pattern, info)
if case:
case_ref_positive = int(case.group(1))
case_ref_negative = int(case.group(2))
case_alt_positive = int(case.group(3)) # read2 is alt read
case_alt_negative = int(case.group(4))
case_ref = case_ref_positive + case_ref_negative
case_alt = case_alt_positive + case_alt_negative
case_pos_depth = case_ref + case_alt
case_var_freq = (case_alt / case_pos_depth) * 100
vcf_info[key].update({
'Case_ref_readsNum': case_ref,
'Case_var_readsNum': case_alt,
'Case_var_Positive_readsNum': case_alt_positive,
'Case_var_Negative_readsNum': case_alt_negative,
'Case_pos_dep': case_pos_depth,
'Case_var_freq': case_var_freq
})
if control:
control_ref_positve = int(control.group(1))
control_ref_negative = int(control.group(2))
control_alt_positve = int(control.group(3))
control_alt_negative = int(control.group(4))
control_ref = control_ref_positve + control_ref_negative
control_alt = control_alt_positve + control_alt_negative
control_pos_depth = control_ref + control_alt
vcf_info[key].update({
'Ctrl_ref_readsNum': control_ref,
'Ctrl_var_readsNum': control_alt,
'Ctrl_var_Positive_readsNum': control_alt_positve,
'Ctrl_var_Negative_readsNum': control_alt_negative,
'Ctrl_pos_dep': control_pos_depth
})
return vcf_info