This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Basic PRS question

Basic question: I have summary stats from a GWAS and want to derive a PRS score from them. I don't have access to the original genotype data used to do the GWAS. Is that possible without genotype information? Is there a tool to do that?

Further info: The data is presented as thus: ID OR OR_se OR_95L OR_95U PVAL
rs309431 0.988757 0.017380 0.953108 1.022454 0.471168

I have looked at this: https://privefl.github.io/bigsnpr/articles/LDpred2.html#computing-ldpred2-scores-genome-wide

What I don't understand is, where do I use the 1000 genomes LD data and where do I "extract" the derived score to run on my own data?

prs
• 1,540 views • link
written by tacrolimus tacrolimus

1 answer

Which summary stats do you have, exactly? Most PRS are constructed from the beta coefficient or Odds Ratio.

• 0 views • link
written by Kevin Blighe Kevin Blighe

I have a set with both

• 0 views • link
written by tacrolimus tacrolimus

Than that is all that you need. Perhaps perform some research (i.e. read publications) on how PRS are calculated.

• 0 views • link
written by Kevin Blighe Kevin Blighe

Thanks Kevin, I have filled out the question further.

• 0 views • link
written by tacrolimus tacrolimus

Log in to answer this question.

More posts like this