Sumitra Muralidhar | Search PubMed

PubMed2026

SAIGE-GPU: accelerating genome- and phenome-wide association studies using GPUs.

MOTIVATION: Genome-wide association studies (GWAS) at biobank scale are computationally intensive, especially for admixed populations requiring robust statistical models. SAIGE is a widely used method for generalized linear mixed-model GWAS but is limited by its CPU-based implementation, making phenome-wide association studies impractical for many research groups. RESULTS: We developed SAIGE-GPU, a GPU-accelerated version of SAIGE that replaces CPU-intensive matrix operations with GPU-optimized kernels. The core innovation is distributing genetic relationship matrix calculations across GPUs and communication layers. Applied to 2068 phenotypes from 635 969 participants in the Million Veteran Program, including diverse and admixed populations, SAIGE-GPU achieved a 5-fold speedup in mixed model fitting on supercomputing infrastructure and cloud platforms. We further optimized the variant association testing step through multi-core and multi-trait parallelization. Deployed on Google Cloud Platform and Azure, the method provided substantial cost and time savings. AVAILABILITY AND IMPLEMENTATION: Source code and binaries are available for download at https://github.com/saigegit/SAIGE/tree/SAIGE-GPU-1.3.3. A code snapshot is archived at Zenodo for reproducibility (DOI: [10.5281/zenodo.17642591]). SAIGE-GPU is available in a containerized format for use across HPC and cloud environments and is implemented in R/C++ and runs on Linux systems.

Genome-Wide Association Study↗

PubMed2025

Primary care providers' perspectives on receiving opportunistic genomic results from a national study: The Million Veteran Program Return Of Actionable Results (MVP-ROAR) Study.

PURPOSE: Patients are increasingly obtaining genetic health information and integrating it into their care with the help of their primary care provider (PCP). However, PCPs may not be adequately prepared to effectively utilize genetic results. Across the Veterans Health Administration health system, the Million Veteran Program Return Of Actionable Results-Familial Hypercholesterolemia (MVP-ROAR-FH) Study clinically confirms and returns genetic results associated with familial hypercholesterolemia (FH), identified in a national biobank program. METHODS: PCPs who received their patient's genetic results through the MVP-ROAR-FH study were invited to participate in semistructured interviews, which explored PCPs' familiarity with FH, how the results affected medical management, and suggestions for process improvement. Interviews were transcribed and analyzed using directed content analysis and constant comparison methods to identify key themes. RESULTS: Interviews with 9 PCPs revealed varied levels of familiarity with genetic testing and FH. Most PCPs did not distinguish FH from common high cholesterol issues and already used similar treatment approaches. Many PCPs did not recall receiving results from the MVP-ROAR-FH study. Alerts in medical records were deemed effective for communicating results. PCPs valued genetics in informing patient care and identifying at-risk family members but noted several implementation barriers, such as additional workload and unclear medical management benefits. Recommendations for improving results disclosure included simplifying the genetic testing report and associated support documents. CONCLUSION: The study represents the first investigation into PCPs' experiences with receiving genetic test results from a biobank linked to a national healthcare system. Results suggest that PCPs generally view genetic testing as beneficial, although they may not significantly alter medical management. PCPs expressed that integrating genetics into routine care may be burdensome and require additional training, which may not be practical. The study underscores the need for accessible genetic information, which could be aided by specialized support roles or different clinical specialties assisting with incorporating genetic results into patient care.

Humans↗