Part 1: The problem Single cell RNA-sequencing (scRNA-seq) is an increasingly popular method for measuring transcriptome-wide gene expression in single cells. However, there are several technical challenges that make analysing scRNA-seq data difficult. One of the biggest hurdles stems from the fact that of the roughly 100,000-300,000 mRNAs in a cell, only 10-40% are captured using current scRNA-seq protocols [1], [2], [3], [4]. As a result, all genes in all cells are undercounted, and lowly expressed genes may be recorded at 0s in the gene expression matrix despite being expressed.