Converting MNP to SNP

Question

[finswimmer]

Hello,

I do Variant Calling with freebayes. For comparison with another vcf from another source I need to split called mnp to snp.

For example freebayes called this:

chr2    21232803    .   TG  CA,CG

But I need this:

chr2    21232803    .   T   C   
chr2    21232804    .   G   A

How can I do this?

fin swimmer

[cpad0112]

Try bcftools norm with m - option.

[finswimmer]

Thanks for your answer.

But this option for bcftools just spilt the multiallelic entry in two seperate lines, resulting in

chr2    21232803    .   TG  CA
chr2    21232803    .   TG  CG

fin swimmer

Answer 1

[cpad0112]

here is the solution to your problem with bcftools and vt and I added an extra record to check the program: (For some reason latest version of vt is crashing on decompose function. Hence bcftools was used. Otherwise, vt alone is fine for this solution)

$ bcftools norm -m -both test.vcf.gz  | vt  decompose_blocksub - -o output.vcf

output ( VCF headers are not shown):

#CHROM  POS ID  REF ALT QUAL    FILTER  INFO
2   21232803    .   T   C   .   .   OLD_CLUMPED=2:21232803:TG/CA
2   21232803    .   T   C   .   .   OLD_CLUMPED=2:21232803:TG/CG
2   21232804    .   G   A   .   .   OLD_CLUMPED=2:21232803:TG/CA
20  763837  .   C   T   .   .   OLD_CLUMPED=20:763837:CA/TG
20  763838  .   A   G   .   .   OLD_CLUMPED=20:763837:CA/TG

input ( VCF headers are not shown):

#CHROM  POS ID  REF ALT QUAL    FILTER  INFO
2   21232803    .   TG  CA,CG
20  763837  .   CA  TG

[finswimmer]

Thanks a lot! That looks the way I like I want it. I try it out the next days.

fin swimmer

Answer 2

[jena]

I would try vcfallelicprimitives from vcflib, especially given that vcflib is from the same author as freebayes.

In fact, the author of freebayes even has a section in the manual about this topic.

[WANG]

nice answer