NIH partners to create SI-ready datasets for models that predict human biology

October 7, 2026

National Institutes Of Health

Objective:

Develop standardized, SI-ready biomedical datasets and resources to support models that predict how cells and biological systems respond to disease and interventions.

Approach:

Partnership: NIH is working with the U.S. Department of Energy, Biohub, and other partners through the Bio Genesis Mission.
Data integration: The effort will draw on existing biomedical datasets, national data infrastructure, NIH repositories, and Common Fund programs developing biological atlases, shared standards, and SI-ready datasets.
Dataset preparation: NIH and Biohub will standardize suitable datasets for model training; the article also identifies a need for measurements across more cell types and conditions and technologies to study cells at greater scale and speed.

Key Findings:

The article describes a planned collaboration and its data-development goals; it reports no completed model or experimental results.
The effort is aligned with the Predicting Living Systems National Science and Technology Challenge.
NIH and Biohub leaders describe virtual-cell models as a potential way to investigate biological questions computationally and prioritize targets and interventions for laboratory and clinical evaluation.

Interpretation:

The initiative aims to make existing and newly generated biomedical data more useful for training predictive models of living systems.

Limitations:

The article provides no specific dataset inventory, timeline, funding details, or technical standards.
It reports no evidence on model performance, validation, or clinical outcomes.