How to filter variants by percentage of supported reads

Question

[Shin Taguchi]

Hi all.

I use VerScan to detect de novo snv. I found the following description in the official manual (https://varscan.sourceforge.net/trio-calling-de-novo-mutations.html), but I have a question.

"In other words, the ideal de novo mutation call will have high depth (>20x) in all three samples, with good support in the child (40-50% of reads for autosomal calls) and no variant-supporting reads in either parent."

What software and what filtering should I use to extract the variants that are supported by 40~50% of the reads? Also, why does it have to be within the 40~50% range?

Thank you.

Answer 1

[Pierre Lindenbaum]

try duphold: https://github.com/brentp/duphold

Also, why does it have to be within the 40~50% range?

a Heterozygous (50% REF/50% ALT) is expected.

[Shin Taguchi]

Thank you for your comment!

Which of duphold's items can be used to filter the above?

a Heterozygous (50% REF/50% ALT) is expected.

Do you mean to extract the site that is most likely not a sequence error?

[Pierre Lindenbaum]

the ideal de novo mutation

you're looking for de novo mutations...