Genome Wide Epistasis Study of On-Statin Cardiovascular Events with Iterative Feature Reduction and Selection

Abstract

Background Predicting risk for major adverse cardiovascular events (MACE) is an evidence-based practice that incorporates lifestyle, history, and other risk factors. Statins reduce risk for MACE by decreasing lipids, but it is difficult to stratify risk following initiation of a statin. Genetic risk determinants for on-statin MACE are low-effect size and impossible to generalize. Our objective was to determine high-level epistatic risk factors for on-statin MACE with GWAS-scale data. Methods Controlled-access data for 5,980 subjects taking a statin collected from Vanderbilt University Medical Center's BioVU were obtained from dbGaP. We used Random Forest Iterative Feature Reduction and Selection (RF-IFRS) to select highly informative genetic and environmental features from a GWAS-scale dataset of patients taking statin medications. Variant-pairs were distilled into overlapping networks and assembled into individual decision trees to provide an interpretable set of variants and associated risk. Results 1,718 cases who suffered MACE and 4,172 controls were obtained from dbGaP. Pathway analysis showed that variants in genes related to vasculogenesis (FDR=0.024), angiogenesis (FDR=0.019), and carotid artery disease (FDR=0.034) were related to risk for on-statin MACE. We identified six gene-variant networks that predicted odds of on-statin MACE. The most elevated risk was found in a small subset of patients carrying variants in COL4A2, TMEM178B, SZT2, and TBXAS1 (OR=4.53, p<0.001). Conclusions The RF-IFRS method is a viable method for interpreting complex "black-box" findings from machine-learning. In this study, it identified epistatic networks that could be applied to risk estimation for on-statin MACE. Further study will seek to replicate these findings in other populations.

Automated quantitative trait locus analysis (AutoQTL)

“…The copyright holder for this preprint this version posted January 15, 2023. ; https://doi.org/10.1101/2023.01.12.523835 doi: bioRxiv preprint [5] and Alzheimer's Disease [6] from simulated and real-world GWAS-scale data. RF-based methodologies have also been designed to identify non-coding variants [7,8] and to predict rare variant pathogenicity [9].…”

“…Even though, to date, autoML packages specifically designed for phenotype to genotype association in omics level data have not been fully explored, there have been several significant and promising attempts. As initial examples, there have been studies that have identified epistatic interactions using random forest (RF) methodologies for major adverse cardiovascular events [5] and Alzheimer's Disease [6] from simulated and real-world GWAS-scale data. RF-based methodologies have also been designed to identify non-coding variants [7,8] and to predict rare variant pathogenicity [9].…”