Question: IGV for WES (Somatic Mutations)
0
gravatar for haiying.kong
3.6 years ago by
haiying.kong290
Germany
haiying.kong290 wrote:

I have identified somatic mutations with MuTect from WES data from paired normal and tumor samples, and am trying to view with IGV the sequences around the genomic location where somatic mutations are identified. I got 2 questions:

(1) Is it absolutely not tolerable if there is even one sequence from normal sample that shows mutation at the genomic location where somatic mutation is identified by MuTect software?

(2) There are noise like mutations at the genomic locations that are not identified as somatic mutation by MuTect. Does this indicate any thing about data quality, or any thing?enter image description here

next-gen • 1.2k views
ADD COMMENTlink modified 3.6 years ago by Chris Miller21k • written 3.6 years ago by haiying.kong290

To answer first question, if your matched normal comes from tissue adjacent to the tumor site, there are chances that it is contaminated with tumor cells and this is why probably you are seeing reads for mutated allele in normal sample. But its okay if you find few reads. As Chris mentioned in his answer, somatic callers are tuned for this kind of noise.

ADD REPLYlink written 3.6 years ago by poisonAlien2.8k
0
gravatar for Chris Miller
3.6 years ago by
Chris Miller21k
Washington University in St. Louis, MO
Chris Miller21k wrote:

1) It depends on how the parameters are set, but in general, somatic callers have to be tolerant of some "noise" in the normal sample that may match the base change in the tumor.

2) Even in the best data, there will be many low-level mutations that are not called as somatic (most rightfully so). What that level of background noise is will depend on a large number of factors (sample quality, library prep, sequencer, etc).

ADD COMMENTlink written 3.6 years ago by Chris Miller21k

enter image description hereThe normal samples are from blood. So unlikely they are contaminated from tumor tissue.

low level of mutations in tumor can be explained by heterogeneity of tumor. But why do we have a lot of very low level of mutations in blood? a lot of times, at the genomic locations where we have very low level of mutations in blood sample do not have any mutations in tumor. Can we explain these as sequencing errors? But if the color of the mutations indicates phred score on IGV, I think some of the mutations have very high phred scores.

IGV figures on publications seem very clean. They do not seem to have much noise.

ADD REPLYlink modified 3.6 years ago • written 3.6 years ago by haiying.kong290
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1922 users visited in the last hour