Check all pathogenics variants

Question

[vpsev3]

Hello,

I have sequenced my genome (WGS 30X) and I have my files (BAM, FASTQ, VCF), what is the most suitable technique to check all the pathogenic variants at once (indel, duplications, known variants etc.)?

I tried to compare my VCF file with ClinVar, without success because none software that I found on Github is working

[Pierre Lindenbaum]

without success because none software that I found on Github is working

wow !

[vpsev3]

On Windows anyway, there are always errors

[Kevin Blighe]

Which errors, and please elaborate further on what you have found. It should be no issue at all to compare your input VCF to a ClinVar VCF.

[German.M.Demidov]

You need special hardware/software for bioinformatics. Usually, it means "not Windows" + at least 32GB of RAM + large HDD to keep all the databases + multi-core processor. Then these tools from github will work.

Answer 1

[Alban Nabla]

Here is a simple method, using only Python:

from cyvcf2 import VCF
cv = VCF("clinvar.vcf.gz") #get file from ClinVar and make sure you have the .tbi file too
gen = VCF("yourGenome.vcf.gz") #get this from your WGS provider with the .tbi file

def compare_vcf(cv, usr):
    variants = {}
    try:
        cvv = next(cv)
        usrv = next(usr)
    except StopIteration:
        return variants
    while True: 
        if cvv.POS > usrv.POS:
            try:
                usrv = next(usr)
            except StopIteration:
                return variants
        if cvv.POS < usrv.POS:
            try:
                cvv = next(cv)
            except StopIteration:
                return variants
        if cvv.POS == usrv.POS:
            if cvv.REF == usrv.REF and cvv.ALT == usrv.ALT:
                variants[cvv.ID] = [cvv.POS, cvv.REF, cvv.ALT, cvv.INFO.get('CLNSIG')]
            try:            
                cvv = next(cv)
            except StopIteration:
                return variants
    return variants

import pandas as pd
chroms = [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,'X','Y'] 
for i in chroms:
    cv_chrom = cv(str(i))
    gen_chrom = gen('chr'+str(i)) #make sure the format is correct for your genome 
    variants = compare_vcf(cv_chrom, gen_chrom)
    output = pd.DataFrame.from_dict(variants, orient='index', columns=['POS', 'REF', 'ALT', 'CLNSIG'])
    output.to_csv('chr'+str(i)+'.csv')