Abstract
Existing computational methods that use single-cell RNA-sequencing (scRNA-seq) for cell fate prediction do not model how cells evolve stochastically and in physical time, nor can they predict how differentiation trajectories are altered by proposed interventions. We introduce PRESCIENT (Potential eneRgy undErlying Single Cell gradIENTs), a generative modeling framework that learns an underlying differentiation landscape from time-series scRNA-seq data. We validate PRESCIENT on an experimental lineage tracing dataset, where we show that PRESCIENT is able to predict the fate biases of progenitor cells in hematopoiesis when accounting for cell proliferation, improving upon the best-performing existing method. We demonstrate how PRESCIENT can simulate trajectories for perturbed cells, recovering the expected effects of known modulators of cell fate in hematopoiesis and pancreatic β cell differentiation. PRESCIENT is able to accommodate complex perturbations of multiple genes, at different time points and from different starting cell populations, and is available at https://github.com/gifford-lab/prescient.
Highlights
Existing computational methods that use single-cell RNA-sequencing for cell fate prediction do not model how cells evolve stochastically and in physical time, nor can they predict how differentiation trajectories are altered by proposed interventions
We show that when accounting for cell proliferation, PRESCIENT outperforms existing methods on predicting cell fate bias
We evaluated on cell fate prediction, which we define to be the task of predicting the clonal fate bias of a given barcoded clone as described by Weinreb et al This is defined as the number of neutrophils divided by the total number of neutrophils and monocytes for that clone (Methods)
Summary
Existing computational methods that use single-cell RNA-sequencing (scRNA-seq) for cell fate prediction do not model how cells evolve stochastically and in physical time, nor can they predict how differentiation trajectories are altered by proposed interventions. Waddington-OT predicts longrange cell–cell probabilistic couplings by reframing the task of inferring cell relationships between population snapshots as an unbalanced optimal transport problem[8] Another method, FateID iteratively builds ensembled cell-type classifiers from labeled terminal cell states[9]. PRESCIENT learns a stochastic, parametric, queryable form of the differentiation landscape via a generative neural network, which enables simulations of high-dimensional trajectories with arbitrary initializations in physical time. This enables simulation of trajectories for cells unobserved during training, including cells with computationally perturbed gene expression profiles, which none of the existing summarization methods or modeling methods are able to do (Fig. 1a). PRESCIENT enables large unbiased in silico perturbation experiments to aid the design of in vitro genetic perturbational screens
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have