The National Institutes of Health (NIH) is coordinating with the U.S. Department of Energy (DOE), Biohub, and other partners to develop data resources for training superintelligence (SI) models that predict how cells and biological systems respond to diseases and potential interventions. Through its Bio Genesis Mission, NIH will bring together existing biomedical datasets, national data infrastructure, and research programs to support the development of resources for the broader scientific community.
Developing predictive models of biology requires large volumes of high-quality data that capture cellular responses to interventions across different cell types and conditions. NIH’s existing biomedical research investments provide a foundation for this effort. These resources include national biomedical repositories maintained by the National Library of Medicine (NLM) and the National Center for Biotechnology Information (NCBI), as well as NIH Common Fund programs developing coordinated biological atlases, shared data standards, and datasets for SI model training.
NIH will work with Biohub to standardize suitable datasets for model training. The effort will require collecting measurements of cellular responses across a wider range of cell types and experimental conditions than currently studied. It will also involve developing technologies capable of observing cells at scales and speeds beyond the capabilities of existing instruments.
The initiative aligns with the Predicting Living Systems National Science and Technology Challenge. It aims to make high-quality biological data more accessible and useful to researchers developing predictive models of living systems. These models could help scientists investigate biological questions computationally, identify potential drug targets and interventions, and prioritize candidates for laboratory and clinical testing.
Click here to read the original press release.
