How Data Engineering Supports Neumora's R&D Programs Using AWS

Though brain diseases account for roughly 20 percent of global deaths, progress in developing effective new therapies has been slow. Neumora Therapeutics, a biopharmaceutical company focused on neurobehavioral disorders, aims to make complex connections within massive volumes of patient genomic data and imaging readouts to drive precision medicine in neuroscience.

Using Amazon Web Services (AWS), Neumora created an integrated data platform that allowed scientists to generate and test hypotheses from complex, multi-class data sets that include imaging, video and audio material, genetic analyses, and other data. Neumora used AWS Batch, a fully managed batch processing service that plans, schedules, and executes batch computing workloads, for data processing in a computational cluster powered by and AWS Lambda, a serverless, event-driven compute service that can run code for virtually any type of application without provisioning or managing servers, to help the company aggregate these pipelines of non-tabular data into a tabular structure in Amazon Simple Storage Service (Amazon S3)—a scalable object storage service—for downstream use. “AWS is the number one company to allow us to process any kind of data set, regardless of the volume, diversity, or type of data. They have all the tools in place,” says Oscar Rodriguez, head of data engineering at Neumora.

For example, the company was able to connect imaging data from more than 10,000 subject MRIs with population-level genetic and biomarker data. These integrated data could then be used to predict behavioral drivers of disability related to mental illness using machine learning (ML).  Using AWS to power automation of complex processing, Neumora reduced the processing time for this project from 90 to 3 days. This workflow fosters collaboration between data engineering, data science, and digital health while prioritizing cost-effectiveness and flexibility of data architecture.

Using AWS infrastructure, Neumora Therapeutics automated complex processing of patient data from heterogenous pipelines to find relationships across data classes, reducing processing time by up to 96 percent.

See how AWS is supporting other leading life science organizations in their quest to improve human health.