RT Journal
A1 Kimmel, Jacob C.
A1 Kelley, David R
T1 Semi-supervised adversarial neural networks for single-cell classification
JF Genome Research 
JO Genome Research 
YR 2021 
FD February 24 
DO 10.1101/gr.268581.120 
SP gr.268581.120 
UL http://genome.cshlp.org/content/early/2021/02/24/gr.268581.120.abstract 
AB Annotating cell identities is a common bottleneck in the analysis of single-cell genomics experiments. Here, we present scNym, a semi-supervised, adversarial neural network that learns to transfer cell identity annotations from one experiment to another. scNym takes advantage of information in both labeled datasets and new, unlabeled datasets to learn rich representations of cell identity that enable effective annotation transfer. We show that scNym effectively transfers annotations across experiments despite biological and technical differences, achieving performance superior to existing methods. We also show that scNym models can synthesize information from multiple training and target datasets to improve performance. In addition to high performance, we show that scNym models are well-calibrated and interpretable with saliency methods.