TorchGWAS 1.0: GPU-accelerated GWAS at scale
Authors:
Open Access
- 7 January 2026
- journal article
- Published by Oxford University Press (OUP) in Bioinformatics Advances
- Vol. 6 (1)
Abstract
Motivation: Imaging, molecular and machine-learning workflows can generate thousands of quantitative phenotypes in one cohort. Repeating genome-wide scans trait by trait creates a computational and output bottleneck. Results: TorchGWAS is a GPU-accelerated framework that uses batched operations for high-throughput, covariate-adjusted linear association testing across large panels of quantitative phenotypes. Across 500 036 allele-harmonized tests, its test statistics agreed with PLINK 2.0, with a Pearson correlation difference from one of 1.3 × 10−12. On an NVIDIA H100 with 80 gigabytes of memory and a 48-core Intel Xeon Gold 6442Y host, median end-to-end runtimes for 4.57 billion associations spanning 8 931 083 variants, 512 phenotypes and 35 365 samples ranged from 28.46 to 58.95 seconds across four genotype formats. These measurements included writing 36.7 gigabytes of binary summary statistics. TorchGWAS supports efficient fixed-effect association screening at scale. Availability and implementation: TorchGWAS is implemented in Python and distributed as a documented source repository at https://github.com/ZhiGroup/TorchGWAS.Keywords
Funding Information
- National Institute on Aging of the National Institutes of Health (U01AG070112)
- National Institutes of Health
This publication has 9 references indexed in Scilit:
- SAIGE-GPU: accelerating genome- and phenome-wide association studies using GPUsBioinformatics, 2026
- Efficient permutation-based genome-wide association studies for normal and skewed phenotypic distributionsBioinformatics, 2022
- rMVP: A Memory-Efficient, Visualization-Enhanced, and Parallel-Accelerated Tool for Genome-Wide Association StudyGenomics, Proteomics & Bioinformatics, 2021
- A resource-efficient tool for mixed model association analysis of large-scale dataNature Genetics, 2019
- Scaling computational genomics to millions of individuals with GPUsGenome Biology, 2019
- The UK Biobank resource with deep phenotyping and genomic dataNature, 2018
- 10 Years of GWAS Discovery: Biology, Function, and TranslationThe American Journal of Human Genetics, 2017
- Second-generation PLINK: rising to the challenge of larger and richer datasetsGigaScience, 2015
- PLINK: A Tool Set for Whole-Genome Association and Population-Based Linkage AnalysesThe American Journal of Human Genetics, 2007