This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Merging GWAS "summary statistics" files for several traits taken from the same samples

I have completed a bunch of GWAS, using the same genotype data each time but with different traits. I now wanted to make a Manhattan plot illustrating all the results, all traits in a single plot. For this need a single "summary statistics" file where each variant only has the lowest p-value obtained among all the GWAS.

What is the best way of accomplishing this? Does anyone know of any tool that might do this? I'm a bit frustrated because I only needed it for a throwaway figure, but I feel like any script I could write for this would take far too long to run.

Thanks in advance!

manhattan gwas genome sumstats snp
• 1,214 views • link
written by antoton • antoton

It would be handy if you provided the format of the summary file(s) you are trying to further summarize; and whether the variants in the files are all the same, and occur in identical orders.

• 0 views • link
written by LChart LChart

Thanks for your reply! I realize now that I didn't give much info above. The files are simple text files with the following columns:

rsid chromosome position A1 A2 pval beta tstat n

Every file has the exact same variants in the exact same order. There are about 7.4M variants. Each file is 580MB. I have a few hundred files but I can subset to those showing significant hits to make it more reasonable.

I'm thinking that if I tabix index each of the files I might be able to do a lookup across all files for each position at a reasonable time-frame? I would gladly hear any opinions or suggestions. Thanks again!

• 0 views • link
written by antoton • antoton

0 answers

No answers yet.

Log in to answer this question.

More posts like this