Protocol for predicting single- and multiple-dose-dependent gene expression using deep generative learning

Variational autoencoders (VAEs) can be used to model the gene expression space of single-cell RNA sequencing (scRNA-seq) data. Here, we present a protocol for using single-cell variational inference of dose response (scVIDR), a VAE designed to model single-cell gene expression for dose-dependent chemical perturbations. We describe steps to access the scVIDR code and data using a containerization application called Docker. We then detail procedures for training the sVIDR model and predicting gene expression. For complete details on the use and execution of this protocol, please refer to Kana et al.¹.