Now published in PNAS, check out my work from UCSF: Unsupervised pattern identification in spatial gene expression atlas reveals mouse brain regions beyond established ontology: https://www.pnas.org/doi/10.1073/pnas.2319804121https://lnkd.in/dEB37EkQ
We combine stability-driven nonnegative matrix factorization with spatial correlation analysis to explore brain regions and ontology in 3D spatial gene expression profiles.
Our findings reveal a gene expression-defined anatomical ontology and interpretable region-specific genetic architecture captured by the marker genes and spatial co-expression networks in the mouse brain.
This shows the power of genomics + machine learning to create biological maps. As the datasets grow and the algorithms improve, I expect this kind of work will lead to a whole generation of medical breakthroughs.
Code available here: https://github.com/abbasilab/osNMFhttps://lnkd.in/dtymd4gC
Great collaboration with Yu Wang, R. Patrick Xian, Alex Lee in collaboration with the Allen Institute for Brain Science's Hongkui Zeng and Bosiljka Tasic and University of California, Berkeley's Bin Yu. Special thanks to Weill Neurohub and and Sandler Program for Breakthrough Biomedical Research for funding our research.