Varscan results annotation

[stqa8350]

Hi,

This is rather a simple problem but I have been unable to find a solution yet.

Varscan (V.2.3.1) snp output has the strand information (eg. tumor_reads1_plus tumor_reads1_minus tumor_reads2_plus tumor_reads2_minus) and the nomenclature is not similar to the vcf headers. This being the requirement for the clinicians, I have chosen to use the varscan snp output (as opposed to using --output-vcf).

However when using annovar as annotation tool (used awk to change some fields to make it compatible to annovar file.input), I still dont get any annotation results (all fields are listed in "summary-snp.annovar.invalid_input" and 0 in exome-summary or variant-function.

Is there a work around for this or any other method to annotate varscan-snp output via annovar ?

Any help is appreciated.

Thanks

Uma

[stqa8350]

Update - found a database solution to this

[dkoboldt]

Uma,

Thanks for following up on this. You can rather easily convert VarScan's native output format to BED-like format to facilitate annotation, e.g.

perl -pe '($chrom, $pos, $ref, $var) = split(/\t/, $_); $start = $pos - 1; $_ = "$chrom\t$start\t$pos\t$ref/$var\n"' varScan.output.snp >varScan.output.snp.bed

We'll also look into expanding VCF output options to included the strand-specific read counts for somatic mutations. You mentioned this is required by clinicians... are their requirements written up somewhere?

Yours,

Dan Koboldt

P.S. VarScan v2.3.1 was released on 8/9/12

[stqa8350]

Hi Dan,

Unfortunately not, its something I intend to develop just to make life simpler.

Thanks for your reply.

Uma