Search

Identification of cell phenotypic states within heterogeneous populations, along with elucidation of their switching dynamics, is a central challenge in modern biology. Conventional single-cell analysis methods typically provide only indirect, static phenotypic readouts. Transmitted light images, on the other hand, provide direct morphological readouts and can be acquired over time to provide a rich data source for dynamic cell phenotypic state identification. Here, we describe an end-to-end deep learning platform, UPSIDE (Unsupervised Phenotypic State IDEntification), for discovering cell states and their dynamics from transmitted light movies. UPSIDE uses the variational auto-encoder architecture to learn latent cell representations, which are then clustered for state identification, decoded for feature interpretation, and linked across movie frames for transition rate inference. Using UPSIDE, we identified distinct blood cell types in a heterogeneous dataset. We then analyzed movies of patient-derived acute myeloid leukemia cells, from which we identified stem-cell associated morphological states as well as the transition rates to and from these states. UPSIDE opens up the use of transmitted light movies for systematic exploration of cell state heterogeneity and dynamics in biology and medicine., Author summary The human body contains hundreds of different cell types, each with distinctive identities and functions. Identifying these cellular identities and functional states is one of the great challenges in contemporary biology. We have addressed this challenge by developing UPSIDE, a machine learning approach for discovering cell types and cell states from time-resolved live cell imaging data, which provides a richly detailed source of information. Key distinguishing features of UPSIDE include the ability to directly learn cell type-defining features from brightfield images without prior knowledge; and an ability to visualize and interpret these machine-learned features to understand why they were chosen. UPSIDE can distinguish different human blood cell types despite their visual similarity and can use morphologic clues to identify and follow cell states. In a disease-relevant example, we used UPSIDE to identify stem-like and more mature cell subpopulations in patient-derived acute myelogenous leukemia cells, and to determine the speed at which these cell types inter-convert. UPSDIDE provides a general-purpose tool for the unbiased identification and analysis of cell types and state transitions in heterogeneous cell populations.