I have exome sequencing data from tumor samples from mice with only one of them has tumor/normal tissue sequence. All the mice share the same genetic background, bred together and i believe they come from the same mother (not 100% sure about the same mother though). I have aligned them sequences using BWA-mem, marked the duplicates, etc.., and now want to call the variants either using VarScan somatic or GATK MuTect2. Is it possible to use this normal tissue sequence as a common Normal tissue for the other tumor sequences from the other mice? or is it better to call the tumors and the normal sequences individually then filter the Germline Variants afterwards from the Tumor samples?
Thanks in advance
•
written
by
tarekzakaria.badr •
1 answer
Perhaps you could try GATK joint variant calls (among your tumor samples)?
While not exactly designed for somatic variant calling, it should increase your sensitivity and help you find recurrent tumor variants.
Log in to answer this question.
-
best pipeline/practice for analyzing WES for germline and somatic variant calling
written by epigene
-
Retain read names in mpileup and VarScan
written by vasily.dobrovolsky •
-
GATK Mutect2 tumor mode
written by cocchi.e89
-
How to detect somatic mutations from blood sample or ctDNA sequencing?
written by Leon
-
the difference between the positions called by somatic caller (MuTect2) and the position called by …
written by Zoey Zou •
-
Is it possible to detect somatic SNVs from C57BL/6 tumor samples WXS without normal tissue control?
written by tarekzakaria.badr •
-
germline calling of tumor-normal paired sample germline by varscan2
written by oghzzang •
-
Somatic and Germline variant calling in ion torrent
written by inayat45shaikh •
-
Calling germline variants from Tumor/Clone pairs to find the somatic mutations of only tumor and on…
written by ivivek_ngs
-
Tool: Varscan: Variant Detection In Massively Parallel Sequencing Data
written by Istvan Albert
Only 1 sample for normal is very little!
However, wherever possible, run T/N pairs , rather than running each of them individually. Normals are there not only to filter the somatic mutations, but also to minimize the technical biases.
https://software.broadinstitute.org/gatk/documentation/article?id=11053
https://gatkforums.broadinstitute.org/gatk/discussion/4462/what-input-files-does-mutect-accept-require
Thanks for your reply. I know it's very little but unfortunately i received the sequences after they have ended the experiments. Following your advice I will then run this normal tissue with the other tumor samples as a Normal/Tumor pair. I also have found SNPS and indels VCF files for mm10 and will try to filter through these also.
Sounds good. Keep in mind that mm8 is a consensus genome, which might be different from the genome of strain of your mice. You could check VCFs of other strains here
https://www.sanger.ac.uk/science/data/mouse-genomes-project