查看文章

Background: In the field of common genetic variant analyses, polygenic risk scores (PRS) have become standard practice for estimating risk on a target test set using GWAS summary statistics of a training set, much like in typical machine learning (ML) applications. The use of genomic annotations that incorporate prior functional information can enhance risk prediction and provide insights into biological risk mechanisms. Here we extract pathway-specific polygenic scores and systematically investigate their potential as a flexible tool for individual prediction, that leverages both multi-dimensional information and the advantages of ML. We apply our method on PGC Schizophrenia (SCZ) and Bipolar Disorder (BIP)“wave 3” data freeze.

Methods: We applied our method on a European ancestry subset of 50,245 cases and 71,185 controls for SCZ and 29,619 cases and 50,914 controls for BIP. Genotypes were put …