You are here
Machine learning models for predicting blood pressure phenotypes by combining multiple polygenic risk scores.
Thursday,2024-07-11 10:26 America/Los_Angeles — ecterry
| Title | Machine learning models for predicting blood pressure phenotypes by combining multiple polygenic risk scores. |
| Publication Type | Journal Article |
| Year of Publication | 2024 |
| Authors | Hrytsenko, Y, Shea, B, Elgart, M, Kurniansyah, N, Lyons, G, Morrison, AC, Carson, AP, Haring, B, Mitchell, BD, Psaty, BM, Jaeger, BC, C Gu, C, Kooperberg, C, Levy, D, Lloyd-Jones, D, Choi, E, Brody, JA, Smith, JA, Rotter, JI, Moll, M, Fornage, M, Simon, N, Castaldi, P, Casanova, R, Chung, R-H, Kaplan, R, Loos, RJF, Kardia, SLR, Rich, SS, Redline, S, Kelly, T, O'Connor, T, Zhao, W, Kim, W, Guo, X, Chen, Y-DIda, Sofer, T |
| Corporate/Institutional Authors | Trans-Omics in Precision Medicine Consortium |
| Journal | Sci Rep |
| Volume | 14 |
| Issue | 1 |
| Pagination | 12436 |
| Date Published | 2024 May 30 |
| ISSN | 2045-2322 |
| Keywords | Blood Pressure, Female, Genetic Predisposition to Disease, Genetic Risk Score, Genome-Wide Association Study, Humans, Hypertension, Machine Learning, Male, Middle Aged, Models, Genetic, Multifactorial Inheritance, Phenotype, Risk Factors |
| Abstract | <p>We construct non-linear machine learning (ML) prediction models for systolic and diastolic blood pressure (SBP, DBP) using demographic and clinical variables and polygenic risk scores (PRSs). We developed a two-model ensemble, consisting of a baseline model, where prediction is based on demographic and clinical variables only, and a genetic model, where we also include PRSs. We evaluate the use of a linear versus a non-linear model at both the baseline and the genetic model levels and assess the improvement in performance when incorporating multiple PRSs. We report the ensemble model's performance as percentage variance explained (PVE) on a held-out test dataset. A non-linear baseline model improved the PVEs from 28.1 to 30.1% (SBP) and 14.3% to 17.4% (DBP) compared with a linear baseline model. Including seven PRSs in the genetic model computed based on the largest available GWAS of SBP/DBP improved the genetic model PVE from 4.8 to 5.1% (SBP) and 4.7 to 5% (DBP) compared to using a single PRS. Adding additional 14 PRSs computed based on two independent GWASs further increased the genetic model PVE to 6.3% (SBP) and 5.7% (DBP). PVE differed across self-reported race/ethnicity groups, with primarily all non-White groups benefitting from the inclusion of additional PRSs. In summary, non-linear ML models improves BP prediction in models incorporating diverse populations.</p> |
| DOI | 10.1038/s41598-024-62945-9 |
| Alternate Journal | Sci Rep |
| PubMed ID | 38816422 |
| PubMed Central ID | PMC11139858 |
| Grant List | K08 HL159318 / HL / NHLBI NIH HHS / United States R01 HL161012 / HL / NHLBI NIH HHS / United States R01HL161012 / HL / NHLBI NIH HHS / United States |
ePub date:
24/06