How the strand bias and R3x6 filter works in illumina platform?

[N3ha247]

I am not able to fully understand the use of the Strand bias filter and RMxN(R3X6) filter in case of somatic variant calling.

For eg.: 1. how the RMXN filter works when a variant is found in the following condition?

   Seq. GGTGGGGGGGG VAF is 34%

   Seq CGCGCGAAAA vaf 25%

   Seq CGCGCGACAGCA vaf 8%

   Seq AGCCAGCCAGCCAGCCAGCCAGCCAGCC vaf 20%

2. How SB is calculated and why does it occur?

   Which is high or low SB? SB -1.000.000 SB -456.122 SB- 0

Can someone please explain to me how these filters are triggered and when to regard or disregard this filter during variant interpretation?​

[tamsen]

Hi there,

The supplementary data in the original paper has a write up of the RMXN filters,

https://academic.oup.com/bioinformatics/article/35/9/1579/5124278#supplementary-data

The wiki also has lots of info (see https://github.com/Illumina/Pisces/wiki). In particular, "How is strand bias calculated? " is in the FAQ. https://github.com/Illumina/Pisces/wiki/Frequently-Asked-Questions . The design docs and Supported Options might also be of interest.

best
Tamsen