Unsupervised deep representation learning enables phenotype discovery for genetic association studies of brain imaging.

Autor: Patel K; McWilliams School of Biomedical Informatics, University of Texas Health Science Center, Houston, TX, 77030, USA., Xie Z; McWilliams School of Biomedical Informatics, University of Texas Health Science Center, Houston, TX, 77030, USA., Yuan H; Department of Computer Science and Engineering, Texas A&M University, College Station, TX, 77843, USA., Islam SMS; McWilliams School of Biomedical Informatics, University of Texas Health Science Center, Houston, TX, 77030, USA., Xie Y; Department of Computer Science and Engineering, Texas A&M University, College Station, TX, 77843, USA., He W; McWilliams School of Biomedical Informatics, University of Texas Health Science Center, Houston, TX, 77030, USA., Zhang W; School of Public Health, University of Texas Health Science Center, Houston, TX, 77030, USA., Gottlieb A; McWilliams School of Biomedical Informatics, University of Texas Health Science Center, Houston, TX, 77030, USA., Chen H; McWilliams School of Biomedical Informatics, University of Texas Health Science Center, Houston, TX, 77030, USA.; School of Public Health, University of Texas Health Science Center, Houston, TX, 77030, USA., Giancardo L; McWilliams School of Biomedical Informatics, University of Texas Health Science Center, Houston, TX, 77030, USA., Knaack A; Department of Neurology and Imaging of Dementia and Aging (IDeA) Laboratory, University of California at Davis, Davis, CA, 95618, USA., Fletcher E; Department of Neurology and Imaging of Dementia and Aging (IDeA) Laboratory, University of California at Davis, Davis, CA, 95618, USA., Fornage M; School of Public Health, University of Texas Health Science Center, Houston, TX, 77030, USA.; McGovern Medical School, University of Texas Health Science Center, Houston, TX, 77030, USA., Ji S; Department of Computer Science and Engineering, Texas A&M University, College Station, TX, 77843, USA., Zhi D; McWilliams School of Biomedical Informatics, University of Texas Health Science Center, Houston, TX, 77030, USA. Degui.Zhi@uth.tmc.edu.
Jazyk: angličtina
Zdroj: Communications biology [Commun Biol] 2024 Apr 05; Vol. 7 (1), pp. 414. Date of Electronic Publication: 2024 Apr 05.
DOI: 10.1038/s42003-024-06096-7
Abstrakt: Understanding the genetic architecture of brain structure is challenging, partly due to difficulties in designing robust, non-biased descriptors of brain morphology. Until recently, brain measures for genome-wide association studies (GWAS) consisted of traditionally expert-defined or software-derived image-derived phenotypes (IDPs) that are often based on theoretical preconceptions or computed from limited amounts of data. Here, we present an approach to derive brain imaging phenotypes using unsupervised deep representation learning. We train a 3-D convolutional autoencoder model with reconstruction loss on 6130 UK Biobank (UKBB) participants' T1 or T2-FLAIR (T2) brain MRIs to create a 128-dimensional representation known as Unsupervised Deep learning derived Imaging Phenotypes (UDIPs). GWAS of these UDIPs in held-out UKBB subjects (n = 22,880 discovery and n = 12,359/11,265 replication cohorts for T1/T2) identified 9457 significant SNPs organized into 97 independent genetic loci of which 60 loci were replicated. Twenty-six loci were not reported in earlier T1 and T2 IDP-based UK Biobank GWAS. We developed a perturbation-based decoder interpretation approach to show that these loci are associated with UDIPs mapped to multiple relevant brain regions. Our results established unsupervised deep learning can derive robust, unbiased, heritable, and interpretable brain imaging phenotypes.
(© 2024. The Author(s).)
Databáze: MEDLINE
Externí odkaz:
Nepřihlášeným uživatelům se plný text nezobrazuje