Slingshot is a novel method for inferring cell lineages and pseudotimes from single-cell transcriptomics data. It combines stable techniques for noisy data with the ability to identify multiple lineages. The method consists of two stages: first, inferring the global lineage structure using a cluster-based minimum spanning tree (MST), and second, inferring pseudotimes using simultaneous principal curves. Slingshot is robust and flexible, allowing for local supervision to identify terminal states and multiple lineages. It outperforms other methods in accuracy and stability, particularly in complex datasets with multiple branching lineages. The method is implemented in the R package slingshot and is compatible with various single-cell data analysis workflows. Slingshot is robust to noise and provides accurate pseudotimes, making it a valuable tool for understanding dynamic gene expression in single-cell data. The method is applicable to a wide range of single-cell datasets and is not restricted by prior knowledge of the number of lineages. It allows for the identification of complex lineage structures and provides a framework for incorporating biological knowledge into the analysis. Slingshot is a modular and flexible method that can be integrated into existing analysis pipelines and is suitable for both supervised and unsupervised analysis. The method is validated on real datasets and simulated studies, demonstrating its effectiveness in accurately inferring lineages and pseudotimes.Slingshot is a novel method for inferring cell lineages and pseudotimes from single-cell transcriptomics data. It combines stable techniques for noisy data with the ability to identify multiple lineages. The method consists of two stages: first, inferring the global lineage structure using a cluster-based minimum spanning tree (MST), and second, inferring pseudotimes using simultaneous principal curves. Slingshot is robust and flexible, allowing for local supervision to identify terminal states and multiple lineages. It outperforms other methods in accuracy and stability, particularly in complex datasets with multiple branching lineages. The method is implemented in the R package slingshot and is compatible with various single-cell data analysis workflows. Slingshot is robust to noise and provides accurate pseudotimes, making it a valuable tool for understanding dynamic gene expression in single-cell data. The method is applicable to a wide range of single-cell datasets and is not restricted by prior knowledge of the number of lineages. It allows for the identification of complex lineage structures and provides a framework for incorporating biological knowledge into the analysis. Slingshot is a modular and flexible method that can be integrated into existing analysis pipelines and is suitable for both supervised and unsupervised analysis. The method is validated on real datasets and simulated studies, demonstrating its effectiveness in accurately inferring lineages and pseudotimes.