“…This paper concerns the application of sparse linear ML to genetic and health information in order to make predictions that could be useful in a clinical setting. Recent work has highlighted that ML, in particular polygenic predictors, have high potential impact in clinical settings [ 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 ], especially for coronary artery disease (CAD) [ 11 , 12 , 22 ]. Additionally, over the past quarter century it has been advocated (e.g., Joint Task Force of the European Society of Cardiology and Other Societies [ 23 ], American College of Cardiology(ACC)/American Heart Association (AHA) [ 24 ], and the Scottish Intercollegiate Guidelines Network [ 25 ]) that physicians should use risk scores based on statistical summaries of biomarkers.…”