SEQanswers

SEQanswers (http://seqanswers.com/forums/index.php)
-   Bioinformatics (http://seqanswers.com/forums/forumdisplay.php?f=18)
-   -   Tumor/Normal comparison tool for VCF? (http://seqanswers.com/forums/showthread.php?t=20478)

NGSNewbie1 05-31-2012 11:56 AM

Tumor/Normal comparison tool for VCF?
 
Sorry for this newbie question but I haven't been able to get an answer after searching around. Is there a program that automagically produces the somatic mutations from inputted tumor and normal VCF files? I'm trying to avoid using a tool like SomaticSniper with BAM files because of size/computation etc. If I have the VCF files, is there an easy way to just select the differences?

Thank you in advance for any guidance!

Bukowski 05-31-2012 01:54 PM

Quote:

Originally Posted by NGSNewbie1 (Post 74885)
Sorry for this newbie question but I haven't been able to get an answer after searching around. Is there a program that automagically produces the somatic mutations from inputted tumor and normal VCF files? I'm trying to avoid using a tool like SomaticSniper with BAM files because of size/computation etc. If I have the VCF files, is there an easy way to just select the differences?

Thank you in advance for any guidance!

The thing is that these tools like SomaticSniper are designed to work on the bam files for a reason. Whilst a subtractive approach with VCFs will work for some somatic calls, most genotypers aren't really geared up for variant detection in heterogenous cancer samples.

NGS is not always speedy work, sometimes it's better just to use the tools than take the shortcuts ;)

ymc 05-31-2012 06:07 PM

Are there any targeted therapies in the work after you did this type of comparison?

ymc 05-31-2012 07:46 PM

An illumina one:

http://www.ncbi.nlm.nih.gov/pubmed/22581179

ymc 05-31-2012 07:53 PM

Another one from Canada

http://www.ncbi.nlm.nih.gov/pubmed/22285562

Bukowski 06-01-2012 12:40 AM

Quote:

Originally Posted by ymc (Post 74929)
Are there any targeted therapies in the work after you did this type of comparison?

If that's directed at me - not in the case of my exome work, but I do work on small, targeted cancer panel development for stratified medicine.

Your also posted Strelka and JointSNVMix - I find the latter somewhat confusing to get to grips with, Strelka is on my to do list. I tend to use VarScan2 for paired comparisons for e.g. copy number. Neither as far as I am concerned answer the question from the OP as they both require BAM files for input.

SkinFlayer 06-01-2012 02:27 PM

Quote:

Originally Posted by NGSNewbie1 (Post 74885)
Sorry for this newbie question but I haven't been able to get an answer after searching around. Is there a program that automagically produces the somatic mutations from inputted tumor and normal VCF files? I'm trying to avoid using a tool like SomaticSniper with BAM files because of size/computation etc. If I have the VCF files, is there an easy way to just select the differences?

Thank you in advance for any guidance!

While I'm uncertain that the results will really be what you're looking for, this can be done using GATK's SelectVariants with the --discordance flag (see http://www.broadinstitute.org/gsa/ga...tVariants.html for details).

Unfortunately, unless you already have GATK up and running, this is not a trivial task (as getting GATK functioning properly can be a bit tricky).

As mentioned by others though, tumor/normal paired variant calling generally seems to be carried out by a program specifically developed for that purpose (VarScan2, SMUG, and MuTect being the ones with which I am familiar).

Sorry there's not a better answer, but I hope this helps a bit.

Jason

shyam_la 06-12-2012 10:10 AM

Quote:

Originally Posted by SkinFlayer (Post 75028)
While I'm uncertain that the results will really be what you're looking for, this can be done using GATK's SelectVariants with the --discordance flag (see http://www.broadinstitute.org/gsa/ga...tVariants.html for details).

Unfortunately, unless you already have GATK up and running, this is not a trivial task (as getting GATK functioning properly can be a bit tricky).

As mentioned by others though, tumor/normal paired variant calling generally seems to be carried out by a program specifically developed for that purpose (VarScan2, SMUG, and MuTect being the ones with which I am familiar).

Sorry there's not a better answer, but I hope this helps a bit.

Jason

Hi,

Do you know how to annotate the output from MuTect? I have 3800 mutation calls and I am stuck..

Baseless 06-14-2012 04:26 AM

@shyam_la

I reformatted mine to something that ensembl variant effect predictor can read and went for that during testing.

And to the others - if someone has advice which of the tools out there comes with a good false discovery rate, i am listening.

shyam_la 06-14-2012 09:46 PM

Quote:

Originally Posted by Baseless (Post 76283)
@shyam_la

I reformatted mine to something that ensembl variant effect predictor can read and went for that during testing.

And to the others - if someone has advice which of the tools out there comes with a good false discovery rate, i am listening.

I discovered SNPEff works as well, with MuTect data. Just need removing the redundant columns...


All times are GMT -8. The time now is 05:02 AM.

Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2021, vBulletin Solutions, Inc.