ML-MAGES enables multivariate genetic association analyses with genes and effect size shrinkage
Genome Research · 2025
DOI: 10.1101/gr.280440.125 ↗Kurz gefasst
Fachartikel in der Zeitschrift Genome Research, 2025. Verfasst von Xiran Liu und 2 weiteren Autor:innen. Beteiligte Einrichtung: Data Science Institute, Brown University, Providence, Rhode Island 02912, USA; xiran_liu1@brown.edu. Hauptthemen laut MEDLINE: Genome-Wide Association Study, Machine Learning. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 61 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIGMS NIH HHS, U.S. National Institutes of Health, Data Science Institute at Brown University.
Abstract (Originalsprache)
A fundamental goal of genetics is to identify which and how genetic variants are associated with a trait, often using the regression results from genome-wide association (GWA) studies. Important methodological challenges account for inflation in GWA effect estimates as well as in investigating more than one trait simultaneously. We leverage machine learning approaches for these two challenges, developing a computationally efficient method called ML-MAGES. First, we shrink the inflation in GWA effect sizes caused by nonindependence among variants using neural networks. We then cluster variant associations among multiple traits via variational inference. We compare the performance of shrinkage via neural networks to regularized regression and fine-mapping, two approaches used for addressing inflated effects but dealing with variants in focal regions of different sizes. Our neural network shrinkage outperforms both methods in approximating the true effect sizes in simulated data. Our infinite mixture clustering approach offers a flexible, data-driven way to distinguish different types of associations—trait-specific, shared across traits, or nonprioritized—among multiple traits based on their regularized effects. Clustering applied to our neural network shrinkage results also produces consistently higher precision and recall for distinguishing gene-level associations in simulations. We demonstrate the application of ML-MAGES on association analyses of two quantitative traits and two binary traits in the UK Biobank. Our identified associated genes from single-trait enrichment tests overlap with those having known relevant biological processes to the traits. Besides trait-specific associations, ML-MAGES identifies several variants with shared multitrait associations, suggesting putative shared genetic architecture.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Data Science Institute, Brown University, Providence, Rhode Island 02912, USA; xiran_liu1@brown.edu.
- Data Science Institute, Brown University, Providence, Rhode Island 02912, USA.
- Microsoft Research, Cambridge, Massachusetts 02142, USA.
- Department of Ecology, Evolution, and Organismal Biology, Brown University, Providence, Rhode Island 02912, USA.
ORCID-Profile: Xiran Liu, Lorin Crawford
Förderung
- NIGMS NIH HHS
- U.S. National Institutes of Health
- Data Science Institute at Brown University
Zitieren
APA
Liu, X., Crawford, L., & Ramachandran, S. (2025). ML-MAGES enables multivariate genetic association analyses with genes and effect size shrinkage. Genome Research. https://doi.org/10.1101/gr.280440.125
Vancouver
Liu X, Crawford L, Ramachandran S. ML-MAGES enables multivariate genetic association analyses with genes and effect size shrinkage. Genome Research. 2025. doi:10.1101/gr.280440.125
BibTeX
@article{Liu2025,
author = {Liu, Xiran and Crawford, Lorin and Ramachandran, Sohini},
title = {ML-MAGES enables multivariate genetic association analyses with genes and effect size shrinkage},
journal = {Genome Research},
year = {2025},
doi = {10.1101/gr.280440.125}
}
EUVIMED