problem with SNPs and Indels detectiong useng WGS data

Question

[Sara]

we have WGS data and looking for SNPs and Indels globally. we found Indels and SNPs using PCR and also when we load the sequencing data on IGV, we can see those events by eye (of course with low frequency). but when we used the following tools we can not detect them in the resulting VCF file:

lofreq , mutect, strelka and bcftool

do you know what could be the problem?

[Pierre Lindenbaum]

you asked many questions on this forum but you validated only one answer. See here

[Sara]

OK thanks for letting me know. do you have any answer for this question?

[Kevin Blighe]

Disagreement between variant callers is expected, and it should also be expected that these [variant callers] will 'miss' (fail to call) genuine variants. For such a question as this, however, you have to explain better the entire pipeline that you've used (including sample preparation, alignment tool and metrics, etc), and show as much relevant code as possible.

Answer 1

[EagleEye]

Hi,

I had this issue while ago for detecting very-rare variants. I was using freebayes and I had to adjust the default parameters (--min-alternate-count 1 --min-alternate-fraction 0.0000001). I am sure when you play with the default parameters you will manage to detect with any tools (you just have to dig in a bit!).