One Question about Varscan

[LiLin]

Hi,I have one question about varscan.

There is one position,I think somatic_status should be Somatic,but the result show it reference.Is there anybody can explain it ?Thank you!

chr position T G 26 1 3.7% T 40 10 20% K Reference 1.0 0.04709966879527449 15 25 4 6

[Jane M]

Hi,

I don't have an answer but something similar:

chr1 16946535 T C 2 16 88,89% C 8 6 42,86% Y Germline 1.0 0.007696151924037949 1 7 1 5

The "somatic p_vlaue" is 0.007 and I chose the threshold as 0.05 so it should be classified as "Somatic"...

[dkoboldt]

Hello LiLin,

There are a few criteria used to determine the somatic status of a variant. In your case, it seems most likely that --somatic-p-value was something lower than 0.47, so the variant did not meet the threshold to be called Somatic. Would you mind providing three things?
1.) The VarScan version you're running
2.) The command you used including all parameters
3.) The normal and tumor pileup/mpileup for that chromosome position (it's fine if you alter the chromosome name or position; we need the depth, bases, and qualities).

Jane, in your example, the variant is not called Somatic because the variant allele frequency is higher in normal (89%) than in tumor (43%).

[LiLin]

Hi,dkoboldt.
1.)VarScan.v2.2.5.jar
2.)Parameters:--min-coverage 10 --min-coverage-normal 10 --min-coverage-tumor 10 --min-var-freq 0.1 --min-freq-for-hom 0.75 --somatic-p-value 0.0001 --min-avg-qual 0 --validation 1
3.)
tumor mpileup:
chr7 126351120 t 50 g,.,gg,.,,,..g,,,,GG..G,,.,,,,...,,...,,,,,g,G..g, 6CDDCCDDDC%DECD%%D;<DC;C%DDDD%CEDCDEDD%DC%4CC;DDB%
normal mpileup:
chr7 126351120 t 27 .,$...,.,,,C.,.,,,.,....,,.. B%EDDDDD%C#DEE%E%DDDEDDCDAD

[Jane M]

Jane, in your example, the variant is not called Somatic because the variant allele frequency is higher in normal (89%) than in tumor (43%).

chr1 16946535 T C 2 16 88,89% C 8 6 42,86% Y Germline 1.0 0.007696151924037949 1 7 1 5

Dan, do you mean that if I had something like

0 100 50 50

it would be classified as germline ?
For me (but I am not biologist), it's a somatic mutation, as

0 100 100 0

would be...
Am I wrong?

[LiLin]

Hello LiLin,

There are a few criteria used to determine the somatic status of a variant. In your case, it seems most likely that --somatic-p-value was something lower than 0.47, so the variant did not meet the threshold to be called Somatic. Would you mind providing three things?
1.) The VarScan version you're running
2.) The command you used including all parameters
3.) The normal and tumor pileup/mpileup for that chromosome position (it's fine if you alter the chromosome name or position; we need the depth, bases, and qualities).

Jane, in your example, the variant is not called Somatic because the variant allele frequency is higher in normal (89%) than in tumor (43%).

Thank you!
Now I set --somatic-p-value 0.05，Then I find it.