Question: VQSR on non-variant positions
gravatar for dmyersturnbull
5.1 years ago by
Stanford University
dmyersturnbull90 wrote:

We have a set of whole-genome Illumina DNA sequences. We're using GATK HaplotypeCaller followed by VQSR, both of which the GATK recommends.

We're interested in only a set of positions that are known to (commonly) contain SNPs, all of which are in dbSNP. For this, we want a list of genotype calls, one for each position in dbSNP, and each annotated with a VQSLOD, its log-odds ratio from VQSR.

Since the positions in question are all known SNP positions, the reference is mostly arbitrary, so it's irrelevant to us whether the genotype call matches the reference. We simply want a list of calls with their confidences, for all positions, including those that were called (by HaplotypeCaller) as homozygous reference. In other words, we want a map f: rsID ----> (GT, VQSLOD).

However, VQSR ignores non-variant positions. How can we get it to score all calls in a set, including the non-variant ones? What procedures or tools have people used to overcome this problem before?


ADD COMMENTlink modified 5.1 years ago • written 5.1 years ago by dmyersturnbull90
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 870 users visited in the last hour