The paper introduces a method called "diffusion pseudotime" (DPT) to robustly estimate the temporal order of differentiation processes in single-cell gene expression data. DPT measures transitions between cells using diffusion-like random walks, allowing for the identification of branching decisions and metastable states. The method is more robust than existing approaches, especially in the presence of noise and outliers, and can handle large-scale datasets without dimensionality reduction. DPT is applied to both simulated and real data, including single-cell qPCR and droplet-based scRNA-seq experiments, demonstrating its ability to identify key regulatory genes and metastable states. Compared to other pseudotime methods like Wanderlust and Monocle, DPT shows higher accuracy in reconstructing temporal orders, particularly in the context of branching lineages. The authors conclude that DPT is a powerful tool for inferring regulatory relationships and understanding differentiation dynamics in single-cell transcriptomics studies.The paper introduces a method called "diffusion pseudotime" (DPT) to robustly estimate the temporal order of differentiation processes in single-cell gene expression data. DPT measures transitions between cells using diffusion-like random walks, allowing for the identification of branching decisions and metastable states. The method is more robust than existing approaches, especially in the presence of noise and outliers, and can handle large-scale datasets without dimensionality reduction. DPT is applied to both simulated and real data, including single-cell qPCR and droplet-based scRNA-seq experiments, demonstrating its ability to identify key regulatory genes and metastable states. Compared to other pseudotime methods like Wanderlust and Monocle, DPT shows higher accuracy in reconstructing temporal orders, particularly in the context of branching lineages. The authors conclude that DPT is a powerful tool for inferring regulatory relationships and understanding differentiation dynamics in single-cell transcriptomics studies.