The paper Mapping the functional network of human cancer through machine learning and pan-cancer...

The paper Mapping the functional network of human cancer through machine learning and pan-cancer proteogenomics has been published in Nature Cancer. Congratulations to co-first authors Zhiao and Jonathan! By applying supervised machine learning to extensive proteomics and RNA sequencing data from 1,194 individuals spanning 11 cancer types, this study constructs a functional network, FunMap, comprising 10,525 protein-coding genes. FunMap achieves unprecedented precision in linking functionally associated genes, surpassing traditional protein-protein interaction maps. Analysis of FunMap identifies functional protein modules, reveals a hierarchical structure linked to cancer hallmarks and clinical phenotypes, provides deeper insights into established cancer drivers and predicts functions for understudied cancer-associated proteins. Additionally, applying graph-neural-network-based deep learning to FunMap uncovers drivers with low mutation frequency. The FunMap Python package is fully open source and available for download from the Python Package Index (https://pypi.org/project/funmap). The source code is hosted on GitHub (https://github.com/bzhanglab/funmap). In addition, a web application, accessible at https://funmap.linkedomics.org/, offers visualization tools to explore gene neighborhoods, dense modules, and the hierarchical organization of this pan-cancer FunMap.