Search

Molecular subtypes, such as defined by The Cancer Genome Atlas (TCGA), delineate a cancer's underlying biology, bringing hope to inform a patient's prognosis and treatment plan. However, most approaches used in the discovery of subtypes are not suitable for assigning subtype labels to new cancer specimens from other studies or clinical trials. Here, we address this barrier by applying five different machine learning approaches to multi-omic data from 8,791 TCGA tumor samples comprising 106 subtypes from 26 different cancer cohorts to build models based upon small numbers of features that can classify new samples into previously defined TCGA molecular subtypes-a step toward molecular subtype application in the clinic. We validate select classifiers using external datasets. Predictive performance and classifier-selected features yield insight into the different machine-learning approaches and genomic data platforms. For each cancer and data type we provide containerized versions of the top-performing models as a public resource., Competing Interests: Declaration of interests A.D.C. receives research support from Bayer and consults for KaryoVerse. W.R. is currently working at Pfizer. I.T., P.C., and V.L. are or were directly or indirectly affiliated with JADBio—Gnosis DA, S.A., which offers the JADBio service commercially. V.F. is an employee and stock option owner of Bluestar Genomics Inc. L.R.R. has received grants from Bayer, Boston Scientific, Exact Sciences, FujiFilm Medical Sciences, Gilead Sciences, GlycoTest, RedHill, and Target PharmaSolutions and serves in a consulting/advising role for AstraZeneca, Bayer, Eisai, Exact Sciences, Gilead Sciences, Global Life Science Consulting, GRAIL, LLC, Hepion, MedEd Design, Medscape, Novartis Venture Fund, QED, RedHill, and The Lynx Group. A.J.L. has consulting relationships with AbbVie, Astra-Zeneca, Bayer, Bio-AI Health, BMS, Caris, Deciphera, Foghorn Therapeutics, GRAIL, GSK, Illumina, Invitae/Archer DX, Iterion Therapeutics, Merck, Novartis, Nucleai, Paige, Pfizer, Regeneron, Roche/Genentech, SpringWorks, Tempus, and ThermoFisher. W.R. and G.G. are co-inventors of a patent application related to lung adenocarcinoma expression subtypes (U.S. Provisional Patent Application No.: 63/293,349). P.W.L. serves on the Scientific Advisory Boards of FOXO Technologies, Inc., and Tagomics, LLC. J.M.S. is a stock owner of Nantomics Inc. V.U. is an employee and stock owner of Bristol Myers Squibb., (Copyright © 2024 The Authors. Published by Elsevier Inc. All rights reserved.)

To identify cancer-associated gene regulatory changes, we generated single-cell chromatin accessibility landscapes across eight tumor types as part of The Cancer Genome Atlas. Tumor chromatin accessibility is strongly influenced by copy number alterations that can be used to identify subclones, yet underlying cis-regulatory landscapes retain cancer type-specific features. Using organ-matched healthy tissues, we identified the "nearest healthy" cell types in diverse cancers, demonstrating that the chromatin signature of basal-like-subtype breast cancer is most similar to secretory-type luminal epithelial cells. Neural network models trained to learn regulatory programs in cancer revealed enrichment of model-prioritized somatic noncoding mutations near cancer-associated genes, suggesting that dispersed, nonrecurrent, noncoding mutations in cancer are functional. Overall, these data and interpretable gene regulatory models for cancer and healthy tissue provide a framework for understanding cancer-specific gene regulation.