Gene set inference from single-cell sequencing data using a hybrid of matrix factorization and variational autoencoders

Recent advances in single-cell RNA sequencing (scRNA-Seq) have driven the simultaneous measurement of the expression of 1,000s of genes in 1,000s of single cells. In principle, these growing data sets now allow us to model the gene sets in biological networks at an unprecedented level of detail, in spite of heterogenous cell populations. To this end we propose an unsupervised deep neural network model that is a novel conditional variational autoencoder (CVA), which utilizes weights as matrix factorizations to obtain gene sets, while class-specific inputs to the latent variable space facilitate a plausible identification of cell types. In essence, this artificial neural network model seamlessly leverages the information of functional gene set inference, experimental batch effect correction, and static gene identification, which we conceptually prove here for three single-cell RNA-Seq datasets and suggest for future single-cell-gene analytics.