Abstract
Single-cell transcriptomics yields ever growing data sets containing RNA expression levels for thousands of genes from up to hundreds of thousands of cells. Common data analysis pipelines include a dimensionality reduction step for visualising the data in two dimensions, most frequently performed using t-distributed stochastic neighbour embedding (t-SNE). It excels at revealing local structure in high-dimensional data, but naive applications often suffer from severe shortcomings, e.g. the global structure of the data is not represented accurately. Here we describe how to circumvent such pitfalls, and explain a protocol for successful exploratory data analysis using t-SNE. They include PCA initialisation, multi-scale similarity kernels, exaggeration, and downsampling-based initialisation for very large data sets. We use published single-cell RNA-seq data sets to demonstrate that this protocol yields superior results compared to the naive application of t-SNE.