Deciphering cell state transitions by logical modeling with single-cell transcriptome data

Cell state transitions are essential for diverse cell fate decision processes. Such transitions are mostly irreversible in natural environments, but they can be reversed under certain conditions. This raises a question on the largely unknown regulatory mechanisms underlying the asymmetric cell state transitions in forward and backward directions. Here, we present scDECIPHER (single-cell dynamic explorer of complex interactions and pathway hierarchies), an integrative framework designed to address this question using single-cell resolution transcriptome data measured over cell state transitions. By applying scDECIPHER to human and mouse cancer samples, we uncover the hidden molecular regulatory mechanisms to overcome drug resistance in cancer, suggesting the potential to control cell state transitions towards desired directions.