samtools mpileup filter SNPs

[Hit]

I used samtools mpileup to call SNPs and indels, followed by vcfutils.pl, as suggested in http://samtools.sourceforge.net/mpileup.shtml

I wonder if there are additional filters that I should use for exome data. I saw in publication people use for example, ">=3 reads support a variation", ">=5% on both strands", "consensus quality >=20". I know "-d3" can be used to assign a minimum number of reads, but can anyone tell me how to set filters for the strand bias? And is the 6th column (QUAL) the "consensus quality" which should be required to be >=20?

Any help is appreciated! Thank you

[dagarfield]

Maybe vcfutils.pl varFilter?

[Hit]

Yes I used varFilter -- I forgot to mention this. I followed the mpileup webpage and used the following two lines:

samtools mpileup -uf ref.fa aln1.bam aln2.bam | bcftools view -bvcg - > var.raw.bcf
bcftools view var.raw.bcf | vcfutils.pl varFilter -D10000 > var.flt.vcf

I have exome data so I used -D10000, but I'm not sure if there are other rules that should be used, especially for those decribed in publications like strand bias (mentioned above).

[dagarfield]

I was thinking specifically of this option

-1 FLOAT min P-value for strand bias (given PV4) [0.0001]

Strand bias is not something I've thought much (at all) about, but this option seems like it might be worth following up.