Dimensionality reduction for visualizing single-cell data using UMAP
By Etienne Becht, Leland McInnes, John Healy, Charles-Antoine Dutertre, I. Kwok, Lai Guan Ng, Florent Ginhoux, Evan W. Newell. 2018
4,708 citations recorded by the source corpus.
Advances in single-cell technologies have enabled high-resolution dissection of tissue composition. Several tools for dimensionality reduction are available to analyze the large number of parameters generated in single-cell studies. Recently, a nonlinear dimensionality-reduction technique, uniform manifold approximation and projection (UMAP), was developed for the analysis of any type of high-dimensional data. Here we apply it to biological data, using three well-characterized mass cytometry and single-cell RNA sequencing datasets. Comparing the performance of UMAP with five other tools, we find that UMAP provides the fastest run times, highest reproducibility and the most meaningful organization of cell clusters. The work highlights the use of UMAP for improved visualization and interpretation of single-cell data.
Subfields: Biophysics, Molecular Biology, Representation & Self-Supervised Learning
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Biophysics
- Dmitry Kobak
- Malte D. Luecken
- Philipp Berens
- Leland McInnes
- Fabian J Theis
- Tieliu Shi
- Yvan Saeys
- John Healy
In Molecular Biology
- Malte D. Luecken
- Fabian J Theis
- Dmitry Kobak
- George C. Linderman
- Philipp Berens
- J. Böhm
- Brian L. Hie
- Kevin Kaichuang Yang
In Representation & Self-Supervised Learning
- Dmitry Kobak
- Philipp Berens
- George C. Linderman
- J. Böhm
- Leland McInnes
- Kevin Kaichuang Yang
- John Healy
- Brian L. Hie
Citations in the corpus
- UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction (2018, methodology): “Based upon preliminary releases of a soware implementation, UMAP has already found widespread use in the elds of bioinformatics [5, 12, 17, 46, 2, 45, 15], materials science [34, 23], and machine learning [14, 20, 21, 24, 19, 47] among others.”
- TISCH: a comprehensive web resource enabling interactive single-cell transcriptome visualization of tumor microenvironment (2020, methodology): “The uniform manifold approximation and projection (UMAP) were utilized to reduce the dimension further and visualize the clustering results (25).”
- Single-Cell RNA-Seq Technologies and Related Computational Data Analysis (2019, background): “Recently, UMAP (uniform manifold approximation and projection) (Becht et al., 2018), and scvis (Ding et al., 2018) were specially developed for reducing the dimensions of scRNA-seq data.”
- A Unifying Perspective on Neighbor Embeddings along the Attraction-Repulsion Spectrum (2020, methodology): “singlecell biology, where researchers routinely use neighbor embedding algorithms for data visualization [2, 16].”
- Pan-cancer single-cell landscape of tumor-infiltrating T cells (2021, methodology): “Second, for the Tex branch, combining Monocle3 (27), RNA velocity, and graph inference based on uniform manifold approximation and projection (UMAP) (28) and STARTRACpairwise transition indexes (pTrans) (14), we inferred twopaths fromnaïve to Tex cell: the first path (P1) going through GZMK Tem cell”
- Comparison Between UMAP and t-SNE for Multiplex-Immunofluorescence Derived Single-Cell Data from Tissue Sections (2019, background): “Among the 119 biopsies included and with a median follow-up of 73 months (4-160), 5-year overall survival was 39.”
- Giotto: a toolbox for integrative analysis and visualization of spatial expression data (2021, methodology): “This interactive viewer allows users to explore the link between cells ’ physical positions and their clustering pattern in the expression space (UMAP or tSNE).”
- An Introduction to Machine Learning (2020, methodology): “24 Dimensionality reduction is also used to visualize the high-dimensional chemical space 33 or as a preprocessing step to improve performance of an ML model.”
- A scalable SCENIC workflow for single-cell gene regulatory network analysis (2020, methodology): “In contrast, the SCENIC UMAP does not appear to show as much clustering; however, it is important to note that this re fl ects a different aspect of the underlying biology.”
- Early computational detection of potential high-risk SARS-CoV-2 variants (2021, methodology): “For unsupervised learning, Uniform Manifold Approximation and Projection (UMAP) was applied, which has been successfully used for analogous problems in biology and is known to render meaningful insights in life science settings [41].”