Computationally efficient whole-genome regression for quantitative and binary traits

Genome-wide association analysis of cohorts with thousands of phenotypes is computationally expensive, particularly when accounting for sample relatedness or population structure. Here we present a novel machine-learning method called REGENIE for fitting a whole-genome regression model for quantitat...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Nature genetics Ročník 53; číslo 7; s. 1097 - 1103
Hlavní autoři: Mbatchou, Joelle, Barnard, Leland, Backman, Joshua, Marcketta, Anthony, Kosmicki, Jack A., Ziyatdinov, Andrey, Benner, Christian, O’Dushlaine, Colm, Barber, Mathew, Boutkov, Boris, Habegger, Lukas, Ferreira, Manuel, Baras, Aris, Reid, Jeffrey, Abecasis, Goncalo, Maxwell, Evan, Marchini, Jonathan
Médium: Journal Article
Jazyk:angličtina
Vydáno: New York Nature Publishing Group US 01.07.2021
Nature Publishing Group
Témata:
45/43
631/114/794
631/208/205
Agriculture
Animal Genetics and Genomics
Association analysis
Biobanks
Biomedical and Life Sciences
Biomedicine
Cancer Research
Case-Control Studies
Chromosomes
Computational biology
Computational Biology - methods
Computer applications
Computer networks
Datasets
Distributed processing
Gene Function
Genome-Wide Association Study - methods
Genomes
Genomics
Genomics - methods
Genotype
Genotypes
Human Genetics
Humans
Learning algorithms
Logistic Models
Machine Learning
Methods
Phenotype
Phenotypes
Population structure
Regression analysis
Regression models
Reproducibility of Results
Statistical analysis
technical-report
United States
United Kingdom > UK
ISSN:1061-4036, 1546-1718, 1546-1718
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Genome-wide association analysis of cohorts with thousands of phenotypes is computationally expensive, particularly when accounting for sample relatedness or population structure. Here we present a novel machine-learning method called REGENIE for fitting a whole-genome regression model for quantitative and binary phenotypes that is substantially faster than alternatives in multi-trait analyses while maintaining statistical efficiency. The method naturally accommodates parallel analysis of multiple phenotypes and requires only local segments of the genotype matrix to be loaded in memory, in contrast to existing alternatives, which must load genome-wide matrices into memory. This results in substantial savings in compute time and memory usage. We introduce a fast, approximate Firth logistic regression test for unbalanced case–control phenotypes. The method is ideally suited to take advantage of distributed computing frameworks. We demonstrate the accuracy and computational benefits of this approach using the UK Biobank dataset with up to 407,746 individuals. REGENIE is a whole-genome regression method based on ridge regression that enables highly parallelized analysis of quantitative and binary traits in biobank-scale data with reduced computational requirements.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1061-4036
1546-1718
1546-1718
DOI:10.1038/s41588-021-00870-7