A0432
Title: Informing two-phase re-sequencing study design via a polygenic risk score convex combination
Authors: Osvaldo Espin-Garcia - University of Western Ontario (Canada) [presenting]
Abstract: The complexity of the genetic architecture of traits makes polygenic risk score (PRS) construction challenging, given that a single PRS method might not summarize the genomic susceptibility of traits comprehensively. For instance, LD-pred-inf and LASSOsum posit contrasting assumptions with respect to the trait underlying genetic architecture: Infinitesimal vs. sparse models. Recent work in a two-phase re-sequencing study design, where only informative subsamples are selected for cost-effective data collection, reduces expenses while preserves the identification of key genetic-disease links. An approach that integrates multiple PRS methods for a two-phase re-sequencing study design is proposed. The proposal solves a convex combination problem aiming to identify the PRS combination that minimizes the mean squared error. In non-edge cases, the resulting combination has the same residuals as a linear regression model with all PRS as covariates, i.e., a residual dependent sampling (RDS). The main advantage of the convex optimization approach is that the resulting PRS combination can be stratified to serve as an auxiliary covariate in maximum likelihood methods, whereas stratification in the model with all PRS as covariates remains unclear. The optimization method is evaluated against alternative RDS designs with single or both PRS methods via Monte Carlo simulations and real data.
