paperKB
coga / coga-kb
Help
Sign in

Chunk #30 — Methods — Summary statistics for PSs

Source
Ancestry deconvolution and partial polygenic score can improve susceptibility predictions in recently admixed individuals.
Embedded
yes

Text

To compute PSs we adopted summary statistics from a collection of studies listed in Supplementary Table 1. Summary statistics were filtered removing palindromic SNPs and those with MAF < 0.01, then clumped applying PLINK 1.935, with parameters: -clump-r2 0.05 -clump-p1 0.5 -clump-p2 1 -clump-kb 1000. The best set of SNPs was selected by determining the best P-value cutoff with PRSice36. Both clumping and cutoff selection were run on the reference set in Supplementary Table 1. Summary statistics for T2D were taken as they are and only clumping was run on the ones for Breast Cancer.