Monocle 2 is a novel computational method for reconstructing complex single-cell trajectories, particularly those with multiple branches. It employs reversed graph embedding (RGE), a machine learning technique that learns a principal graph to describe the trajectory in a lower-dimensional space. This method does not require prior knowledge about the number of cell fates or branch points, making it highly flexible and robust. Monocle 2 was applied to two studies of blood development, revealing that mutations in key lineage transcription factors can divert cells to alternative fates. The method outperforms existing tools in terms of accuracy and robustness, demonstrating its effectiveness in resolving complex branching processes in single-cell data.Monocle 2 is a novel computational method for reconstructing complex single-cell trajectories, particularly those with multiple branches. It employs reversed graph embedding (RGE), a machine learning technique that learns a principal graph to describe the trajectory in a lower-dimensional space. This method does not require prior knowledge about the number of cell fates or branch points, making it highly flexible and robust. Monocle 2 was applied to two studies of blood development, revealing that mutations in key lineage transcription factors can divert cells to alternative fates. The method outperforms existing tools in terms of accuracy and robustness, demonstrating its effectiveness in resolving complex branching processes in single-cell data.