Roche Accelerates Personalized Healthcare R&D While Reducing Costs Using AWS HealthOmics

Roche, developing personalized medicines, had built an on-premises genomics platform (REDCAP) to analyze multimodal health data, but it faced scaling challenges, required several hours to run workflows, and consumed expensive compute resources. Roche adopted AWS HealthOmics for production genomics pipelines, using Nextflow and Workflow Description Language for whole-genome sequencing analyses, with the service automatically selecting optimal compute resources and letting researchers access private and Ready2Run workflows securely worldwide. This cut cancer research analysis time from 1 year to 3 months (an 80% reduction), delivered 90% storage cost savings through automatic S3 lifecycle archiving, and achieved 40% compute cost savings by eliminating idle capacity after jobs complete; Roche has since extended the approach to multimodal analytics including medical imaging via AWS HealthImaging. Senior Infrastructure Architect Victor Sankarlingam: "We use AWS HealthOmics to cut the time to complete analysis for cancer research from 1 year to 3 months."

AWS is Amazon's cloud platform, used across pharma for genomics compute, SageMaker/Bedrock model training, and healthcare-and-life-sciences-specific services like Amazon HealthOmics.

This sits in the Build stage of the AI use-case lifecycle: Named examples from the foundation-model labs and platforms pharma teams use to actually construct, train and generate the model or application.