1
|
Elgart M, Lyons G, Romero-Brufau S, Kurniansyah N, Brody JA, Guo X, Lin HJ, Raffield L, Gao Y, Chen H, de Vries P, Lloyd-Jones DM, Lange LA, Peloso GM, Fornage M, Rotter JI, Rich SS, Morrison AC, Psaty BM, Levy D, Redline S, Sofer T. Non-linear machine learning models incorporating SNPs and PRS improve polygenic prediction in diverse human populations. Commun Biol 2022; 5:856. [PMID: 35995843 PMCID: PMC9395509 DOI: 10.1038/s42003-022-03812-z] [Citation(s) in RCA: 27] [Impact Index Per Article: 13.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/19/2021] [Accepted: 08/05/2022] [Indexed: 01/03/2023] Open
Abstract
Polygenic risk scores (PRS) are commonly used to quantify the inherited susceptibility for a trait, yet they fail to account for non-linear and interaction effects between single nucleotide polymorphisms (SNPs). We address this via a machine learning approach, validated in nine complex phenotypes in a multi-ancestry population. We use an ensemble method of SNP selection followed by gradient boosted trees (XGBoost) to allow for non-linearities and interaction effects. We compare our results to the standard, linear PRS model developed using PRSice, LDpred2, and lassosum2. Combining a PRS as a feature in an XGBoost model results in a relative increase in the percentage variance explained compared to the standard linear PRS model by 22% for height, 27% for HDL cholesterol, 43% for body mass index, 50% for sleep duration, 58% for systolic blood pressure, 64% for total cholesterol, 66% for triglycerides, 77% for LDL cholesterol, and 100% for diastolic blood pressure. Multi-ancestry trained models perform similarly to specific racial/ethnic group trained models and are consistently superior to the standard linear PRS models. This work demonstrates an effective method to account for non-linearities and interaction effects in genetics-based prediction models.
Collapse
Affiliation(s)
- Michael Elgart
- Division of Sleep and Circadian Disorders, Brigham and Women's Hospital, Boston, MA, USA.
- Department of Medicine, Harvard Medical School, Boston, MA, USA.
| | - Genevieve Lyons
- Division of Sleep and Circadian Disorders, Brigham and Women's Hospital, Boston, MA, USA
- Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, MA, USA
| | - Santiago Romero-Brufau
- Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, MA, USA
- Department of Medicine, Mayo Clinic, Rochester, MN, USA
| | - Nuzulul Kurniansyah
- Division of Sleep and Circadian Disorders, Brigham and Women's Hospital, Boston, MA, USA
| | - Jennifer A Brody
- Cardiovascular Health Research Unit, Department of Medicine, University of Washington, Seattle, WA, USA
| | - Xiuqing Guo
- The Institute for Translational Genomics and Population Sciences, Department of Pediatrics, The Lundquist Institute for Biomedical Innovation at Harbor-UCLA Medical Center, Torrance, CA, USA
| | - Henry J Lin
- The Institute for Translational Genomics and Population Sciences, Department of Pediatrics, The Lundquist Institute for Biomedical Innovation at Harbor-UCLA Medical Center, Torrance, CA, USA
| | - Laura Raffield
- Department of Genetics, University of North Carolina, Chapel Hill, NC, USA
| | - Yan Gao
- The Jackson Heart Study, University of Mississippi Medical Center, Jackson, MS, USA
| | - Han Chen
- Human Genetics Center, Department of Epidemiology, Human Genetics, and Environmental Sciences, School of Public Health, The University of Texas Health Science Center at Houston, Houston, TX, USA
- Center for Precision Health, School of Biomedical Informatics, The University of Texas Health Science Center at Houston, Houston, TX, USA
| | - Paul de Vries
- Human Genetics Center, Department of Epidemiology, Human Genetics, and Environmental Sciences, School of Public Health, The University of Texas Health Science Center at Houston, Houston, TX, USA
| | | | - Leslie A Lange
- Department of Medicine, University of Colorado Denver, Anschutz Medical Campus, Aurora, CO, USA
| | - Gina M Peloso
- Department of Biostatistics, Boston University School of Public Health, Boston, MA, USA
| | - Myriam Fornage
- Human Genetics Center, Department of Epidemiology, Human Genetics, and Environmental Sciences, School of Public Health, The University of Texas Health Science Center at Houston, Houston, TX, USA
- Brown Foundation Institute of Molecular Medicine, McGovern Medical School, University of Texas Health Science Center at Houston, Houston, TX, USA
| | - Jerome I Rotter
- The Institute for Translational Genomics and Population Sciences, Department of Pediatrics, The Lundquist Institute for Biomedical Innovation at Harbor-UCLA Medical Center, Torrance, CA, USA
| | - Stephen S Rich
- Center for Public Health Genomics, University of Virginia School of Medicine, Charlottesville, VA, USA
| | - Alanna C Morrison
- Human Genetics Center, Department of Epidemiology, Human Genetics, and Environmental Sciences, School of Public Health, The University of Texas Health Science Center at Houston, Houston, TX, USA
| | - Bruce M Psaty
- Cardiovascular Health Research Unit, Departments of Medicine, Epidemiology, and Health Services, University of Washington, Seattle, WA, USA
| | - Daniel Levy
- The Population Sciences Branch of the National Heart, Lung and Blood Institute, Bethesda, MD, USA
- The Framingham Heart Study, Framingham, MA, USA
| | - Susan Redline
- Division of Sleep and Circadian Disorders, Brigham and Women's Hospital, Boston, MA, USA
- Department of Medicine, Harvard Medical School, Boston, MA, USA
| | - Tamar Sofer
- Division of Sleep and Circadian Disorders, Brigham and Women's Hospital, Boston, MA, USA.
- Department of Medicine, Harvard Medical School, Boston, MA, USA.
- Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, MA, USA.
| |
Collapse
|
2
|
Clark KC, Kwitek AE. Multi-Omic Approaches to Identify Genetic Factors in Metabolic Syndrome. Compr Physiol 2021; 12:3045-3084. [PMID: 34964118 PMCID: PMC9373910 DOI: 10.1002/cphy.c210010] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/06/2022]
Abstract
Metabolic syndrome (MetS) is a highly heritable disease and a major public health burden worldwide. MetS diagnosis criteria are met by the simultaneous presence of any three of the following: high triglycerides, low HDL/high LDL cholesterol, insulin resistance, hypertension, and central obesity. These diseases act synergistically in people suffering from MetS and dramatically increase risk of morbidity and mortality due to stroke and cardiovascular disease, as well as certain cancers. Each of these component features is itself a complex disease, as is MetS. As a genetically complex disease, genetic risk factors for MetS are numerous, but not very powerful individually, often requiring specific environmental stressors for the disease to manifest. When taken together, all sequence variants that contribute to MetS disease risk explain only a fraction of the heritable variance, suggesting additional, novel loci have yet to be discovered. In this article, we will give a brief overview on the genetic concepts needed to interpret genome-wide association studies (GWAS) and quantitative trait locus (QTL) data, summarize the state of the field of MetS physiological genomics, and to introduce tools and resources that can be used by the physiologist to integrate genomics into their own research on MetS and any of its component features. There is a wealth of phenotypic and molecular data in animal models and humans that can be leveraged as outlined in this article. Integrating these multi-omic QTL data for complex diseases such as MetS provides a means to unravel the pathways and mechanisms leading to complex disease and promise for novel treatments. © 2022 American Physiological Society. Compr Physiol 12:1-40, 2022.
Collapse
Affiliation(s)
- Karen C Clark
- Department of Physiology, Medical College of Wisconsin, Milwaukee, Wisconsin, USA
| | - Anne E Kwitek
- Department of Physiology, Medical College of Wisconsin, Milwaukee, Wisconsin, USA
| |
Collapse
|
3
|
Okazaki A, Horpaopan S, Zhang Q, Randesi M, Ott J. Genotype Pattern Mining for Pairs of Interacting Variants Underlying Digenic Traits. Genes (Basel) 2021; 12:1160. [PMID: 34440333 PMCID: PMC8391494 DOI: 10.3390/genes12081160] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/24/2021] [Revised: 07/23/2021] [Accepted: 07/27/2021] [Indexed: 12/15/2022] Open
Abstract
Some genetic diseases ("digenic traits") are due to the interaction between two DNA variants, which presumably reflects biochemical interactions. For example, certain forms of Retinitis Pigmentosa, a type of blindness, occur in the presence of two mutant variants, one each in the ROM1 and RDS genes, while the occurrence of only one such variant results in a normal phenotype. Detecting variant pairs underlying digenic traits by standard genetic methods is difficult and is downright impossible when individual variants alone have minimal effects. Frequent pattern mining (FPM) methods are known to detect patterns of items. We make use of FPM approaches to find pairs of genotypes (from different variants) that can discriminate between cases and controls. Our method is based on genotype patterns of length two, and permutation testing allows assigning p-values to genotype patterns, where the null hypothesis refers to equal pattern frequencies in cases and controls. We compare different interaction search approaches and their properties on the basis of published datasets. Our implementation of FPM to case-control studies is freely available.
Collapse
Affiliation(s)
- Atsuko Okazaki
- Department of Diagnostics and Therapeutics of Intractable Diseases, Juntendo University, Bunkyo-ku, Tokyo 113-8421, Japan;
- Laboratory of Statistical Genetics, Rockefeller University, New York, NY 10065, USA
| | - Sukanya Horpaopan
- Department of Anatomy, Faculty of Medical Science, Naresuan University, Phitsanulok 65000, Thailand;
| | - Qingrun Zhang
- Department of Mathematics and Statistics, University of Calgary, Calgary, AB T2N 1N4, Canada;
| | - Matthew Randesi
- Laboratory of the Biology of Addictive Diseases, Rockefeller University, New York, NY 10065, USA;
| | - Jurg Ott
- Laboratory of Statistical Genetics, Rockefeller University, New York, NY 10065, USA
| |
Collapse
|