Varscan 2 output

Question

[senowinski]

I have a couple of questions regarding the Varscan 2 output/options. The first is regarding the --min-coverage option. I used the following:

--mpileup 1 --tumor-purity 1 --min-coverage 20 --min-var-freq 0.01 --somatic-p-value 0.00001

I was expecting to see variants only called in the tumour only if they were 20 reads over that bp/indel. As I am not seeing this using --min-coverage 20, could someone explain to me what this option means? And what is the best method to determine if a variant is real or not?

We have fresh frozen tumor with matched normal that we sequenced to a depth of ~100x. Are there other parameters I should take into consideration too? We are comparing two different subtypes (invasive and non-invasive) and would like to identify those variants that are transmitted between these two groups.

Thanks!

Answer 1

[poisonAlien]

--min-coverage - Minimum coverage in normal and tumor to call variant [8]

As it says, you need 20 reads in both normal as well as in tumor for a site to be considered for variant calling. I think your --somatic-p-value is too stringent. We generally use 0.05 and it works good enough to detect confident calls. What do you mean by "I am not seeing this"? Are you observing zero Somatic calls?

Regarding variant being real or not, there is no way to tell this just by looking at the data. You can just detect confident calls which are possible candidates (assuming pure sample of course), for further validation process (Sanger, etc).