JIND

Joint Integration and Discrimination for Automated Single-Cell Annotation. JIND is a framework for automated cell-type identification based on neural networks. It directly learns a low-dimensional representation (latent code) inwhich cell-types can be reliably determined. To account for batch effects, JIND performs a novel asymmetric alignment in which the transcriptomic profileof unseen cells is mapped onto the previously learned latent space, hence avoiding the need of retraining the model whenever a new dataset becomes available. JIND also learns cell-type-specific confidence thresholds to identify and reject cells that cannot be reliably classified. We show on datasets with and without batch effects that JIND classifies cells more accurately than previously proposed methods while rejecting only a small proportion of cells. Moreover, JIND batch alignment is parallelizable, being more than five or six times faster than Seurat integration.