BABEL: using deep learning to translate between single-cell datasets

Abstract

Recent advances in sequencing and barcoding technologies have enabled researchers to simultaneously profile gene expression, chromatin accessibility, and/or protein levels in single cells. However, these multiomic techniques often pose technical and financial barriers that limit their practicality. Kevin Wu and colleagues recently developed BABEL, a deep learning algorithm that can effectively translate between transcriptomic and chromatin profiles in single cells, thereby enabling researchers to perform multiomic analyses from an individual dataset.