CORAL in Nature Genetics unifies multiscale spatial omics via graph learning
elhamazizi · x · 2026-11-06
James Zou's group and collaborators published CORAL (Cross-scale spatial Omics Registration and Analysis via graph Learning) in Nature Genetics — a probabilistic deep generative model for integrating spatial omics data.
- Problem: spatial assays trade off spatial resolution against molecular coverage, and cross-modal alignment is hard
- Method: graph attention jointly models high-resolution low-coverage datasets with low-resolution genome-wide ones, integrating transcriptomic, proteomic, metabolomic and epigenomic modalities into unified representations
- Published online November 5, 2026, with authors including Siyu He, Matthew Bieniosek and Jingtian Zhou
More from Apps
- Impasto turns any photo into a palette-knife painting with zero neural networks — CatAstro_Piyush · 2026-11-06
- Self-hosted Excalidraw app adds MCP so Claude can draw diagrams and read them back — KindlyLeadership7978 · 2026-11-06
- Musecases now catalogs 600+ real-world Meta Muse AI agent use cases — thisiskp_ · 2026-11-06
- Amazon lets you text Alexa your shopping list, with photos, to place orders — The Verge AI · 2026-11-06
- Viktor gives its AI employee an email address — just CC it to delegate work — hey_abusiddik · 2026-11-06
- AI-run Pixel News Network hits Amazon Fire TV just 6 weeks after launch — adamamcbride · 2026-11-06