Computing linkage disequilibrium aware genome embeddings using autoencoders
Author:
Taş Gizem1ORCID, Westerdijk Timo2, Postma Eric3, , van Rheenen Wouter, Bakker Mark K, van Eijk Kristel R, Kooyman Maarten, Al Khleifat Ahmad, Iacoangeli Alfredo, Ticozzi Nicola, Cooper-Knock Johnathan, Gromicho Marta, Chandran Siddharthan, Morrison Karen E, Shaw Pamela J, Hardy John, Sendtner Michael, Meyer Thomas, Başak Nazli, Fogh Isabella, Chiò Adriano, Calvo Andrea, Pupillo Elisabetta, Logroscino Giancarlo, Gotkine Marc, Vourc’h Patrick, Corcia Philippe, Couratier Philippe, Millecamps Stèphanie, Salachas François, Mora Pardina Jesus S, Rojas-García Ricardo, Dion Patrick, Ross Jay P, Ludolph Albert C, Weishaupt Jochen H, Freischmidt Axel, Bensimon Gilbert, Tittmann Lukas, Lieb Wolfgang, Franke Andre, Ripke Stephan, Whiteman David C, Olsen Catherine M, Uitterlinden Andre G, Hofman Albert, Amouyel Philippe, Traynor Bryan, Singleton Adrew B, Neto Miguel Mitne, Cauchi Ruben J, Ophoff Roel A, van Deerlin Vivianna M, Grosskreutz Julian, Graff Caroline, Brylev Lev, Rogelj Boris, Koritnik Blaž, Zidar Janez, Stević Zorica, Drory Vivian, Povedano Monica, Blair Ian P, Kiernan Matthew C, Nicholson Garth A, Henders Anjali K, de Carvalho Mamede, Pinto Susana, Petri Susanne, Weber Markus, Rouleau Guy A, Silani Vincenzo, Glass Jonathan, Brown Robert H, Landers John E, Shaw Christopher E, Andersen Peter M, Garton Fleur C, McRae Allan F, McLaughlin Russell L, Hardiman Orla, Kenna Kevin P, Wray Naomi R, Al-Chalabi Ammar, Van Damme Philip, van den Berg Leonard H, Veldink Jan H, Veldink Jan H2ORCID, Schönhuth Alexander4ORCID, Balvert Marleen1ORCID
Affiliation:
1. Department of Econometrics and Operations Research, Tilburg University , Tilburg 5037AB, The Netherlands 2. Department of Neurology, University Medical Center Utrecht , Utrecht 3584CX, The Netherlands 3. Department of Cognitive Science and Artificial Intelligence, Tilburg University , Tilburg 5037AB, The Netherlands 4. Faculty of Technology, Bielefeld University , Bielefeld 33615, Germany
Abstract
Abstract
Motivation
The completion of the genome has paved the way for genome-wide association studies (GWAS), which explained certain proportions of heritability. GWAS are not optimally suited to detect non-linear effects in disease risk, possibly hidden in non-additive interactions (epistasis). Alternative methods for epistasis detection using, e.g. deep neural networks (DNNs) are currently under active development. However, DNNs are constrained by finite computational resources, which can be rapidly depleted due to increasing complexity with the sheer size of the genome. Besides, the curse of dimensionality complicates the task of capturing meaningful genetic patterns for DNNs; therefore necessitates dimensionality reduction.
Results
We propose a method to compress single nucleotide polymorphism (SNP) data, while leveraging the linkage disequilibrium (LD) structure and preserving potential epistasis. This method involves clustering correlated SNPs into haplotype blocks and training per-block autoencoders to learn a compressed representation of the block’s genetic content. We provide an adjustable autoencoder design to accommodate diverse blocks and bypass extensive hyperparameter tuning. We applied this method to genotyping data from Project MinE, and achieved 99% average test reconstruction accuracy—i.e. minimal information loss—while compressing the input to nearly 10% of the original size. We demonstrate that haplotype-block based autoencoders outperform linear Principal Component Analysis (PCA) by approximately 3% chromosome-wide accuracy of reconstructed variants. To the extent of our knowledge, our approach is the first to simultaneously leverage haplotype structure and DNNs for dimensionality reduction of genetic data.
Availability and implementation
Data are available for academic use through Project MinE at https://www.projectmine.com/research/data-sharing/, contingent upon terms and requirements specified by the source studies. Code is available at https://github.com/gizem-tas/haploblock-autoencoders.
Funder
Dutch ALS Foundation
Publisher
Oxford University Press (OUP)
|
|