Project Grant R35GM166045

The National Institute of General Medical Sciences awarded the University of Illinois $395,194 on July 21, 2026, under the Biomedical Research and Research Training program (CFDA 93.859) to develop interpretable machine learning models for decoding cellular dynamics in single-cell and spatial omics datasets.

The project will design multimodal machine learning approaches that integrate transcriptomics, imaging data, and text-based knowledge to disentangle gene programs and cellular factors that drive morphological features and abnormal cellular behaviors. The work will leverage perturbation datasets to model how genetic perturbations, cell-intrinsic properties, and environmental changes influence cell-state transitions, enabling large-scale therapeutic screening. Researchers will generate synthetic spatial transcriptomics data and create disentangled cell representations across diverse single-cell study contexts to improve interpretability of how distinct biological factors influence cellular behavior.

The University of Illinois at Chicago is the recipient. Performance occurs in Chicago, Illinois. The award period runs through April 30, 2031.

This is a Project Grant, the assistance type under which NIGMS supports investigator-initiated basic research that advances understanding of biological processes and lays the foundation for disease diagnosis, treatment, and prevention.