Can Deep Learning Improve Genomic Prediction of Complex Human Traits?. Issue 3 (31st August 2018)

• Record Type: Journal Article
• Authors: Bellot, Pau; de los Campos, Gustavo; Pérez-Enciso, Miguel
• Is Part Of: Genetics. Volume 210:Issue 3(2018)
• Page Start: 809
• Page End: 819
• Publication Date: 2018-08-31
• Subjects: Convolutional Neural Networks -- complex traits -- deep learning -- genomic prediction -- Multilayer Perceptrons -- UK Biobank -- whole-genome -- Genomic Prediction regressions -- GenPred; Genetics -- Periodicals; 576.5
• DOI: 10.1534/genetics.118.301298
• Languages: English
• ISSNs: 0016-6731

Abstract

Abstract: The current excitement around artificial intelligence and the renewed interest in "deep learning" (DL) have been applied to the genetic analysis of complex traits; however, the performance of DL for genomic prediction of complex... The genetic analysis of complex traits does not escape the current excitement around artificial intelligence, including a renewed interest in "deep learning" (DL) techniques such as Multilayer Perceptrons (MLPs) and Convolutional Neural Networks (CNNs). However, the performance of DL for genomic prediction of complex human traits has not been comprehensively tested. To provide an evaluation of MLPs and CNNs, we used data from distantly related white Caucasian individuals ( n ∼100k individuals, m ∼500k SNPs, and k = 1000) of the interim release of the UK Biobank. We analyzed a total of five phenotypes: height, bone heel mineral density, body mass index, systolic blood pressure, and waist–hip ratio, with genomic heritabilities ranging from ∼0.20 to 0.70. After hyperparameter optimization using a genetic algorithm, we considered several configurations, from shallow to deep learners, and compared the predictive performance of MLPs and CNNs with that of Bayesian linear regressions across sets of SNPs (from 10k to 50k) that were preselected using single-marker regression analyses. For height, a highly heritable phenotype, all methods performed similarly, although CNNs were slightly but consistently worse. For the rest of the phenotypes, the performance of some CNNs was comparable or slightly better than linear methods. Performance of MLPs was highly dependent on SNP set and phenotype. In all, over the range of traits evaluated in this study, CNN performance was competitive to linear models, but we did not find any case where DL outperformed the linear model by a sizable margin. We suggest that more research is needed to adapt CNN methodology, originally motivated by image analysis, to genetic-based problems in order for CNNs to be competitive with linear models.