Project Grant R35GM162431

Duke University's Office of Research Administration is executing a $444,125 Project Grant awarded by the National Institute of General Medical Sciences (NIGMS) under the Biomedical Research and Research Training program (CFDA 93.859). The award, effective May 1, 2026, through February 28, 2031, supports the development of advanced computational methods and bioinformatics software for analyzing spatial transcriptomics data at sub-cellular resolution with high gene coverage. The research team will deliver machine learning methods for supervised cell type mapping by integrating reference single-cell transcriptomics data, with comprehensive performance evaluation using single-cell RNA-based simulations and validation by tissue histology experts.

Additionally, the project will produce unsupervised machine learning approaches for cell state discovery to identify spatial organization and regulatory mechanisms beyond traditional cell type classification, along with analytical methodologies to correct diffusion noise and identify direct cell-cell interactions in high-resolution spatial data. These computational tools and analytical approaches will be developed and validated across diverse tissue contexts, including tumor microenvironments and inflammatory tissues, to enable researchers to fully leverage the capabilities of emerging spatial transcriptomics technologies for genome-wide expression profiling.