This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Converting genotype data to SNP calls

Hi all,

I have genotype data obtained through an Infinium array. The genotype is either AA, AB, BB. I would like to find the SNP (i.e. A to C, T to G, etc.) at each site for each sample. I have annotation data for each SNP site giving the A and B alleles. Therefore, I'm thinking of converting this genotype data into SNP calls following these simple rules:

  • If genotype is AA, don't report anything
  • If genotype is AB or BB, report a SNP from A to B

My question is does this algorithm make sense, biologically?

Thanks.

snp illumina genotype
• 3,551 views • link
updated by Jorge Amigo Jorge Amigo • written by novice novice

1 answer

I don't exactly see why would you want to do so, but "biologically speaking" your algorithm "makes sense" if you are willing to create just a catalog of variant sites from your original data.

  • AA is an homozygous reference genotype, so you could consider it as a non variant site. you could therefore avoid writing A to A or A>A, since the reference allele is the only one present on both copies of that diploid organism you sure are working with.
  • both AB and BB imply a change from the reference allele, heterozygous and homozygous respectively and they could both be written as A to B or A>B, but you would lose the zygosity information. you'll have to be sure that the downstream analyses you'll perform on this new data don't need or benefit from that information.
• 0 views • link
written by Jorge Amigo Jorge Amigo

Awesome.

The reason I want to do this is to produce a list of SNPs that I can compare to other SNPs obtained from WES and WGS (in vcf format).

• 0 views • link
written by novice novice

I see. you'll probably find it easy to grep by "B" your Infinium array data, cut the chromosome and position columns, sort those positions and then filter your vcf files using that chr-pos list similarly to grep B infiniumdata.txt | cut -f1,2 | sort -k1,1 -k2,2n | bcftools view -R - WESorWGSdata.vcf, assuming chr and pos columns in your infinium data are 1 and 2 respectively. grepping vcf file instead of region filtering it through bcftools should also work since you're only interested in SNP positions.

• 0 views • link
written by Jorge Amigo Jorge Amigo

Log in to answer this question.

More posts like this