To date, most applications calculate PRS from a subset of the genetic markers after pruning out single nucleotide polymorphisms (SNPs) in LD and applying a P-value threshold to GWAS summary statistics3. Although this approach has advantages in terms of computational and conceptual simplicity, and has been used to predict genetic liability across a broad phenotypic spectrum, recent studies have shown that this conventional method for PRS construction discards information and limits prediction accuracy4. More sophisticated Bayesian polygenic prediction methods that rely on GWAS summary statistics, including LDpred4 and the normal-mixture model recently developed5,6, can incorporate genome-wide markers and accommodate varying genetic architectures, and thus have enhanced performance and flexibility. However, the type of prior on SNP effect sizes used in these methods, known as discrete mixture priors, imposes daunting computational challenges and may result in inaccurate adjustment for local LD patterns.
