Abstract
Clinically-actionable risk models may be significantly improved by including genetic information via polygenic risk scores (PRSs). Several attempts to determine the degree of PRS improvement have reached inconclusive answers, likely due to their focus on a few diseases and statistics. This investigation aims to address this uncertainty by comprehensively analyzing 23 diseases within the UK Biobank. Our results show that including the PRS to a base model containing age and sex significantly improves prediction accuracy, as measured by ROC curves, in a majority 21 of 23 diseases. Furthermore, we find over 100 instances in which lifestyle-modifications or medications reduce disease risk for individuals in the top PRS quintile significantly more than for individuals in the bottom PRS quintile. Lastly, we find minimal confounding by social and demographic factors, and describe an effective methodology for PRS creation. Together, these results indicate that PRSs carry far greater clinical impact than previously known.
Submitted Version
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call