How to check for mutations common across all samples?

[ronton]

I have .bam, .vcf, and ANNOVAR .csv files for several samples.

Is there a straightforward way to sort or view mutations that are the most common or present in several samples?

I was thinking of trying to sort the variant location columns in ascending order and line them up for all the samples, but then how would I create a list of the most common variants?

Any ideas are appreciated, thank you.

[blakeoft]

I had to do this once. It's not a perfect solution, but check out this thread: Get common lines from multiple files. Specifically, look at what the user Radoulov posts. Unfortunately, the solution is just provided and not explained so it might be hard to tweak it.

You'll probably have to take out a few columns of your vcf though. If you just want the chrom, pos and alt, you could do:

awk '{print $1,$2,$5}' file.vcf

I'm sure there are some other things that you would have to do with the output if it's not exactly what you want, but I think it can be done with a few unix commands.

Edit: One more thing. If you wanted to do several pairwise comparisons (which you probably don't), look into using unix's comm.

[ronton]

Thank you for the help, I was able to do something with what you suggested.

It looks like many of the samples have tons of the same mutations so I'm not sure how fruitful this little test was. Perhaps I can refine it.

[blakeoft]

You might also filter out the variants in dbsnp, or at least do them separately.

[swbarnes2]

Programs like BEDTools can give you the intersection of multiple .vcf files.