Modeling gene interactions in polygenic prediction via geometric deep learning
Genome Research · 2025
DOI: 10.1101/gr.279694.124 ↗Kurz gefasst
Fachartikel in der Zeitschrift Genome Research, 2025. Verfasst von Han Li und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Mathematical Sciences and LPMC, Nankai University, Tianjin, 300071, China. Hauptthemen laut MEDLINE: Deep Learning, Multifactorial Inheritance, Epistasis, Genetic, Models, Genetic. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 94 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: National Natural Science Foundation of China, National Key Research and Development Program of China, New Cornerstone Science Foundation.
Abstract (Originalsprache)
Polygenic risk score (PRS) is a widely used approach for predicting individuals’ genetic risk of complex diseases, playing a pivotal role in advancing precision medicine. Traditional PRS methods, predominantly following a linear structure, often fall short in capturing the intricate relationships between genotype and phenotype. In this study, we present PRS-Net, an interpretable geometric deep learning–based framework that effectively models the nonlinearity of biological systems for enhanced disease prediction and biological discovery. PRS-Net begins by deconvoluting the genome-wide PRS at the single-gene resolution and then explicitly encapsulates gene–gene interactions leveraging a graph neural network (GNN) for genetic risk prediction, enabling a systematic characterization of molecular interplay underpinning diseases. An attentive readout module is introduced to facilitate model interpretation. Extensive tests across multiple complex traits and diseases demonstrate the superior prediction performance of PRS-Net compared with a wide range of conventional PRS methods. The interpretability of PRS-Net further enhances the identification of disease-relevant genes and gene programs. PRS-Net provides a potent tool for concurrent genetic risk prediction and biological discovery for complex diseases.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- School of Mathematical Sciences and LPMC, Nankai University, Tianjin, 300071, China.
- Institute for Interdisciplinary Information Sciences, Tsinghua University, Beijing, 100084, China.
- School of Engineering, Research Center for Industries of the Future, Westlake University, Hangzhou, 310030, Zhejiang, China; zengjy@westlake.edu.cn mpsnyder@stanford.edu sai.zhang@ufl.edu.
- Department of Genetics, Center for Genomics and Personalized Medicine, Stanford University School of Medicine, Stanford, California 94304, USA; zengjy@westlake.edu.cn mpsnyder@stanford.edu sai.zhang@ufl.edu.
- Department of Epidemiology, University of Florida, Gainesville, Florida 32603, USA; zengjy@westlake.edu.cn mpsnyder@stanford.edu sai.zhang@ufl.edu.
- Departments of Biostatistics & Biomedical Engineering, UF Genetics Institute, University of Florida, Gainesville, Florida 32603, USA.
ORCID-Profile: Han Li
Förderung
- National Natural Science Foundation of China
- National Key Research and Development Program of China
- New Cornerstone Science Foundation
- Research Center for Industries of the Future
- Westlake Education Foundation
Zitieren
APA
Li, H., Zeng, J., Snyder, M. P., & Zhang, S. (2025). Modeling gene interactions in polygenic prediction via geometric deep learning. Genome Research. https://doi.org/10.1101/gr.279694.124
Vancouver
Li H, Zeng J, Snyder MP, Zhang S. Modeling gene interactions in polygenic prediction via geometric deep learning. Genome Research. 2025. doi:10.1101/gr.279694.124
BibTeX
@article{Li2025,
author = {Li, Han and Zeng, Jianyang and Snyder, Michael P and Zhang, Sai},
title = {Modeling gene interactions in polygenic prediction via geometric deep learning},
journal = {Genome Research},
year = {2025},
doi = {10.1101/gr.279694.124}
}
EUVIMED