Decipher: A computational pipeline to extract context-specific mechanistic insights from single-cell profiles

Abstract

The advent of single-cell profiling technologies has revolutionized our understanding of the cellular and molecular states that underpin disease. However, current computational tools struggle to recover both known and novel mechanistic insights at distinct layers of biological regulation. Here, we present Decipher, a novel computational pipeline that builds integrated cell signalling networks from single-cell profiles in a context-specific, data-driven manner and identifies the key cellular and molecular events that drive disease. We benchmarked Decipher against existing tools and found it could recover known, experimentally determined cytokine signalling pathways, whilst maintaining the flexibility to detect novel pathways and context-specific effects. Notably, Decipher produces global cell-to-cell signalling maps that are interpretable. We utilised Decipher to unveil the cellular and molecular mechanisms driving a novel population of inflammatory monocytes enriched with interferon stimulated genes that is markedly increased in frequency following secondary immunization with the Pfizer-BioNTech COVID-19 mRNA vaccine. Finally, we employed Decipher to interrogate regulon profiles from covid-19 patients with mild versus severe disease, and we found that progression to severe disease was associated with a loss of interferon signalling transcription factors (Irf7, Irf9, STAT1, STAT2) and a gain of factors that drive inflammation and cellular stress responses (NFkB, HIF-1a, ATF3, ATF4). Taken together, our findings demonstrate that Decipher can decode signalling pathways and report on ligand-receptor mediated transcription factor-target gene networks that underlie processes in homeostasis, disease, and cellular responses to therapies. We present Decipher as an invaluable new tool for the discovery of novel therapeutic targets and the development of new medicines.