The art of using t-SNE for single-cell transcriptomics

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.

Embedding to Reference t-SNE Space Addresses Batch Effects in Single-Cell Classification

“…In our experiments, we used a mixture of two Gaussian kernels with perplexity values of 50 and 500. We note that a similar formulation of multi-scale kernels was proposed in [15], and we found the resulting embeddings are visually very similar to those obtained with the approach described above (not shown for brevity).…”

“…Larger values may result in a more globally consistent visualisations, preserving distances on a large scale and organizing clusters in a more meaningful way. Larger values of perplexity can lead to merging of multiple small clusters, thus obscuring local aspects of the data [15].…”

“…The proposed solution is implemented using a mapping of new data to an existing t-SNE visualization. While the utility of such an algorithm was already hinted at in recent publication [15], here we provide its practical and theoretically-grounded implementation. Considering the abundance of recent publications on batch effect removal, we present surprising evidence that a computationally more direct and principled embedding procedure solves the batch effects problem when constructing interpretable visualizations from different data sources.…”

A quantitative framework for evaluating single-cell data structure preservation by dimensionality reduction techniques

“…User-defined parameters for unsupervised algorithms often present themselves as "black-box" knobs with unknown consequences. Tuning these parameters can be a daunting task for the single-cell analyst, but is known to be crucial to algorithm performance (Belkina et al, 2018;Kobak and Berens, 2019;Tsuyuzaki et al, 2019).…”

“…Because these techniques condense cell features in the native space to a small number of latent dimensions for visualization, lost information can result in exaggerated or dampened cell-cell similarity. Furthermore, depending on input data and user-defined parameters, the structure of resulting embeddings can vary greatly, potentially altering biological interpretation (Kobak and Berens, 2019).…”