TOGGLE identifies fate and function process within a single type of mature cell through single-cell transcriptomics

Almost inevitably, mature cells meet their demise through specific processes of phenotypic change such as apoptosis. Although single-cell RNA sequencing (scRNA-seq) technology can reconstruct processes like life cycles and disease progression, identifying specific types of cell death in dying cells remains challenging due to the complexity of related signaling pathways and marker genes. To address this, our study introduces a fate prediction method, TOGGLE, which reconstructs complete programmed cell fates from untracked data. Based on a corollary of Takens’ theorem, TOGGLE enables analysis of highly similar programmed cell death data with indistinct fate boundaries. This approach circumvents the need for live sample tracking and serial assessments. In hematopoiesis and reprogramming, TOGGLE reconstructs full fate biases and predicts cell differentiation from early development to maturity. In brain tissue from a murine model of cerebral infarction, TOGGLE predicts the fate trajectories of neurons undergoing post-infarction programmed cell death (especially ferroptosis) and identifies genes expressed differentially between dying and healthy neurons. Subsequently, functional transitions and differences were discussed using cells from the heart.