Varscan output contains dots

[Irina Pulyakhina]

Hi guys,

I'm using VarScan v 2.3.7 to call SNPs in Illumina Hiseq2000 RNA-Seq paired-end sample:

> samtools mpileup -B -Q 0 -f reference.fa $CONTROL.bam > sample.mpileup
> java -jar VarScan mpileup2cns $sample.mpileup --min-coverage 10 --min-reads2 3 --strand-filter 0 --output-vcf 1 > sample.vcf

What I get a lot is columns that contain dots at a position of the variance:
chr1 155979352 . A . . PASS ADP=464;WT=1;HET=0;HOM=0;NC=0 GT:GQ:SDPP:RD:AD:FREQ:PVAL:RBQ:ABQ:RDF:RDR:ADF:ADR 0/0:666:536:464:448:14:3.03%:5.523E-5:37:34:315:133:14:0

When I check the reads in UCSC, I can see 448 A (reference allele), 14 G (alternative allele) and no other nucleotides. I'm not sure why, instead of reporting "G", Varscan reports a dot as a variance.

Does anyone have an explanation for this?

Thanks in advance!

Cheers,
Irina

[yueluo]

3% variant frequency does not pass the default minimum variant frequency for Varscan, which is 0.2(20%) for "mpileup2cns". You can change the threshold with '--min-var-freq' option.

[nnutter]

This question was cross-posted on Biostars.