Empowering rare variant burden-based gene-trait association studies via optimized computational predictor choice

Abstract

Computational predictors can help interpret pathogenicity of human genetic variants, especially for the majority of variants where no experimental data are available. However, because we lack a high-quality unbiased test set, identifying the best-performing predictors remains a challenge. To address this issue, we evaluated missense variant effect predictors using genotypes and traits from a prospective cohort. We considered 139 gene-trait combinations with rare-variant burden association based on at least one of four systematic studies using phenotypes and whole-exome sequences from ~200K UK Biobank participants. Using an evaluation set of 35,525 rare missense variants and the relevant associated traits, we assessed the correlation of participants' traits with scores derived from 20 computational variant effect predictors. We found that two predictors—VARITY and REVEL—outperformed all others according to multiple performance measures. We expect that this study will help in selecting variant effect predictors, for both research and clinical purposes, while providing an unbiased benchmarking strategy that can be applied to additional cohorts and predictors.

Updated benchmarking of variant effect predictors using deep mutational scanning

“…Prediction performance on case–control disease studies would also not be reliant on existing clinical labels, but would greatly reduce the diversity of variants tested (McInnes et al , 2019; Wu et al , 2021). This approach can also be scaled up and applied to multiple relevant gene‐trait combinations (preprint: Kuang et al , 2022).…”

Interpreting protein variant effects with computational predictors and deep mutational scanning

“…Envision has been found to accurately predict effect magnitude in DMS datasets ( Reeb et al, 2020 ), although our group found that Envision produced an average performance by using a different set of DMS data ( Livesey and Marsh, 2020 ). The ability of VARITY to predict DMS data has yet to be assessed, although a study using gene-trait combinations found it to have excellent predictive performance ( Kuang et al, 2021 ).…”