From Harmony to Discord: Multicellular Coordination in Tissues and Its Rewiring in Cancer

Tissue function emerges from coordinated interactions among diverse cell types, but how these interactions are structured and rewired in disease remains unclear. In a recent study, Shi and colleagues introduce CoVarNet, a computational framework that maps reproducible multicellular modules (CM) across 35 human tissues using single-cell and spatial transcriptomics. These CMs, spanning immune, stromal, and endothelial cells, exhibit functional organization across tissue systems and dynamically respond to biological transitions such as aging and menopause. Importantly, cancer progression is marked by a breakdown of tissue-specific CMs and the emergence of a convergent cancer-associated ecosystem, cCM02. This rewiring reflects a fundamental reorganization of tissue architecture during malignancy and provides new opportunities for diagnostics and therapeutic targeting. The study signifies a conceptual advance from cell-centric to ecosystem-level biology and offers a generalizable framework for integrating multimodal data to dissect tissue-level coordination. In this issue, we discuss how CoVarNet redefines our understanding of tissue organization, its translational implications in oncology, and unresolved questions in modular tissue biology. This article is part of a special series: Driving Cancer Discoveries with Computational Research, Data Science, and Machine Learning/AI.