Context-aware deconvolution of cell–cell communication with Tensor-cell2cell

• DOI: 10.1038/s41467-022-31369-2
• Publication Date: 2022-06-27T10:07:15Z
• Authors: Erick Armingol; Hratch M. Baghdas...; Cameron Martino; Araceli Perez-Lopez; Caitlin Aamodt; Rob Knight; Nathan E. Lewis
• Keywords: DECIPHER

Abstract Cell interactions determine phenotypes, and intercellular communication is shaped by cellular contexts such as disease state, organismal life stage, tissue microenvironment. Single-cell technologies measure the molecules mediating cell–cell communication, emerging computational tools can exploit these data to decipher communication. However, current methods either disregard context or rely on simple pairwise comparisons between samples, thus limiting ability complex across multiple time points, levels of severity, spatial contexts. Here we present Tensor-cell2cell, an unsupervised method using tensor decomposition, which deciphers context-driven simultaneously accounting for stages, states, locations cells. To do so, Tensor-cell2cell uncovers patterns associated with different phenotypic states determined unique combinations cell types ligand-receptor pairs. As such, robustly improves upon extends analytical capabilities existing tools. We show identify modules distinct processes (e.g., participating pairs) linked severities Coronavirus Disease 2019 Autism Spectrum Disorder. Thus, introduce effective easy-to-use strategy understanding diverse conditions.