Deep generative model deciphers derailed trajectories in acute myeloid leukemia

Single-cell genomics has the potential to map cell states and their dynamics in an unbiased way in response to perturbations like disease. However, elucidating the cell-state transitions from healthy to disease requires analyzing data from perturbed samples jointly with unperturbed reference samples. Existing methods for integrating and jointly visualizing single-cell datasets from distinct contexts tend to remove key biological differences or do not correctly harmonize shared mechanisms. We present Decipher, a model that combines variational autoencoders with deep exponential families to reconstruct derailed trajectories (). Decipher jointly represents normal and perturbed single-cell RNA-seq datasets, revealing shared and disrupted dynamics. It further introduces a novel approach to visualize data, without the need for methods such as UMAP or TSNE. We demonstrate Decipher on data from acute myeloid leukemia patient bone marrow specimens, showing that it successfully characterizes the divergence from normal hematopoiesis and identifies transcriptional programs that become disrupted in each patient when they acquire NPM1 driver mutations. ### Competing Interest Statement D.P. is on the scientific advisory board of Insitro. R.L.L. is on the supervisory board of Qiagen and on the board of directors of Ajax Therapeutics, for which he receives compensation and equity support. He is or has recently been a scientific advisor to Imago, Mission Bio, Syndax. Zentalis, Ajax, Bakx, Auron, Prelude, C4 Therapeutics and Isoplexis for which he receives equity support. He has research support from Ajax and AbbVie, consulted for Janssen, and received honoraria from Astra Zeneca and Kura for invited lectures.