MuSIC Maps: Composing Cell Maps to Explore Disease

Get Program Updates

Immunofluorescence image of the target protein stained green, the nuclei are represented in blue, and the cytoskeleton are in red. The photo is credited to the Human Protein Atlas.

Advances in biology continue to increase our understanding of the human body by uncovering new knowledge about one of the essential units in our anatomy, cells. Cells range in size and function with complex structures that remain largely uncharted, making it difficult for scientists to understand the impact of diseases. There is a need to map human cells across the body to gain new insights into proteins, which play a key role in cell structure and function. Mapping cell structure can offer a deeper understanding of the complex nature of human cells and have significant implications for researchers as they study cancer and other chronic diseases.

To better understand the human body at the cellular level, researchers supported by the NIH Common Fund’s Bridge to Artificial Intelligence (Bridge2AI) program combined advanced imaging with artificial intelligence tools—including GPT-4—to study how proteins interact and function. Focusing on human bone cancer cells, the researchers analyzed more than 5,100 proteins to create detailed MultiScale Integrated Cell (MuSIC) maps. These maps revealed 275 unique protein assemblies—clusters of proteins working together on specific tasks—and uncovered previously unknown functions for 975 proteins. Some of these newly identified roles include involvement in RNA processing, which plays a key part in cell function.

Using MuSIC maps, the research team pinpointed multiple protein assemblies that are often mutated in pediatric cancers and uncovered over 100 new proteins linked to cancer development. These newly identified links provide researchers with valuable new targets to explore in their efforts to slow pediatric cancer progression.

For more information about the cell maps, visit the Cell Maps for AI (CM4AI) project page or the MuSIC Maps website.

Reference: Schaffer LV, Hu M, Qian G, Moon KM, Pal A, Soni N, Latham AP, Pontano Vaites L, Tsai D, Mattson NM, Licon K, Bachelder R, Cesnik A, Gaur I, Le T, Leineweber W, Palar A, Pulido E, Qin Y, Zhao X, Churas C, Lenkiewicz J, Chen J, Ono K, Pratt D, Zage P, Echeverria I, Sali A, Harper JW, Gygi SP, Foster LJ, Huttlin EL, Lundberg E, Ideker T. Multimodal cell maps as a foundation for structural and functional genomics. Nature. 2025 Apr 9. doi: 10.1038/s41586-025-08878-3. Epub ahead of print. PMID: 40205054.