A Prism Vote method for individualized risk prediction of traits in genotype data of Multi-population

• Published in: PLoS genetics Vol. 18; no. 10; p. e1010443
• Main Authors: Xia, Xiaoxuan, Zhang, Yexian, Sun, Rui, Wei, Yingying, Li, Qi, Chong, Marc Ka Chun, Wu, William Ka Kei, Zee, Benny Chung-Ying, Tang, Hua, Wang, Maggie Haitian
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 27.10.2022; Public Library of Science (PLoS)
• Subjects: Bayes Theorem; Bayesian analysis; Biology and Life Sciences; Disease; Eigenvalues; Engineering and Technology; Genetic research; Genetic structure; Genetic susceptibility; Genome-Wide Association Study; Genomes; Genomics - methods; Genotype; Genotype & phenotype; Genotypes; Health aspects; Health risk assessment; Identification and classification; Medicine and Health Sciences; Methods; Minority & ethnic groups; Models, Genetic; Phenotype; Physical Sciences; Polymorphism, Single Nucleotide; Predictions; Research and Analysis Methods; Sample variance; Simulation; Subpopulations; Hong Kong
• ISSN: 1553-7404; 1553-7390; 1553-7404
• DOI: 10.1371/journal.pgen.1010443

Multi-population cohorts offer unprecedented opportunities for profiling disease risk in large samples, however, heterogeneous risk effects underlying complex traits across populations make integrative prediction challenging. In this study, we propose a novel Bayesian probability framework, the Prism Vote (PV), to construct risk predictions in heterogeneous genetic data. The PV views the trait of an individual as a composite risk from subpopulations, in which stratum-specific predictors can be formed in data of more homogeneous genetic structure. Since each individual is described by a composition of subpopulation memberships, the framework enables individualized risk characterization. Simulations demonstrated that the PV framework applied with alternative prediction methods significantly improved prediction accuracy in mixed and admixed populations. The advantage of PV enlarges as genetic heterogeneity and sample size increase. In two real genome-wide association data consists of multiple populations, we showed that the framework considerably enhanced prediction accuracy of the linear mixed model in five-group cross validations. The proposed method offers a new aspect to analyze individual's disease risk and improve accuracy for predicting complex traits in genotype data.