Deep learning analysis of single‐cell data in empowering clinical implementation

Abstract

Recent advances in single-cell sequencing technologies enable the characterization of cellular heterogeneity and biological processes in complex diseases. This provides unprecedented opportunities to understand disease pathology at a level that allows mechanistic classification and development of precision therapeutic strategies. Extensive research has been performed in clinical studies at the single-cell level. 1 In addition, emerging deep learning (DL) technologies hold great potential in modeling large-volume and highly heterogeneous single-cell data by using sophisticated architectures, such as artificial neural networks, 2 for translational and clinical purpose. 3 In this commentary, we focus on the DL analysis of singlecell data in empowering the clinical implementation of personalized medicine. DEVELOPING DIAGNOSIS METHODSSingle-cell technology (e.g., single-cell RNA sequencing [scRNA-seq]) for characterizing diseased cell populations Anjun Ma and Juexin Wang contributed equally to this study.