The art of using t-SNE for single-cell transcriptomics
Preprint
Abstract
Single-cell transcriptomics yields ever growing data sets containing RNA expression levels for thousands of genes from up to millions 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 develop a protocol for creating more faithful t-SNE visualisations. It includes PCA initialisation, a high learning rate, and multi-scale similarity kernels; for very large data sets, we additionally use exaggeration and downsampling-based initialisation. 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.Keywords
All Related Versions
- Published version: Nature Communications, 10 (1), 1.
This publication has 43 references indexed in Scilit:
- GPU accelerated t-distributed stochastic neighbor embeddingJournal of Parallel and Distributed Computing, 2019
- The single-cell transcriptional landscape of mammalian organogenesisNature, 2019
- M3Drop: dropout-based feature selection for scRNASeqBioinformatics, 2018
- Dimensionality reduction for visualizing single-cell data using UMAPNature Biotechnology, 2018
- Automated optimized parameters for t-distributed stochastic neighbor embedding improve visualization and allow analysis of large datasetsbioRxiv, 2018
- Identification of cell types in a mouse brain single-cell atlas using low sampling coverageBMC Biology, 2018
- Generalizable and Scalable Visualization of Single-Cell Data Using Neural NetworksCell Systems, 2018
- Electrophysiological, transcriptomic and morphologic profiling of single neurons using Patch-seqNature Biotechnology, 2015
- Mapping the stereotyped behaviour of freely moving fruit fliesJournal of The Royal Society Interface, 2014
- viSNE enables visualization of high dimensional single-cell data and reveals phenotypic heterogeneity of leukemiaNature Biotechnology, 2013