python操作VCF文件

安装

pip install PyVCF

使用 

#如果有多个样本，可以这样访问最后两列信息，例如双样本，0为normal，1为tumor

record.samples[0]

对vcf进行筛选，并重新生成vcf

切记用完一定要close()

import vcf

filaname1='xxx.vcf'
filename2='xxx.vcf'
tumor_idx=1
vcf_reader = vcf.Reader(open(filename1, 'r'))
vcf_writer = vcf.Writer(open(filename2,'w'),vcf_reader)
for record in vcf_reader:
    if len(record.FILTER) == 0: #pass
        af = record.samples[tumor_idx]['AF']
        if af >= 0.01:
            vcf_writer.write_record(record)
vcf_writer.close()

也可以读取vcf.gz

import vcf 
 
vcf_name = "xx.vcf.gz"
vcf_reader = vcf.Reader(filename = vcf_name)
for record in vcf_reader:
    print(record.CHROM)  ##str
    print(record.POS)    ##int
    print(record.REF)    ##str
    print(record.ALT)    ##list
    print(record.INFO)   ##dict
    print(record.FORMAT)   ##str