The DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index.
Initialization is critical for preserving global data structure in both t-SNE and UMAP
Open accessAbstract
1Institute for Ophthalmic Research, University of Tübingen, Tübingen, Germany. 2Applied Mathematics Program, Yale University, New Haven, CT, USA. ✉e-mail: dmitry.kobak@uni-tuebingen.de; george.linderman@yale.edu One of the most ubiquitous analysis tools in single-cell transcriptomics and cytometry is t-distributed stochastic neighbor embedding (t-SNE)1, which is used to visualize individual cells as points on a two-dimensional scatterplot such that similar cells are positioned close together2. A related algorithm, called uniform manifold approximation and projection (UMAP)3, has attracted substantial attention in the single-cell community4. In Nature Biotechnology, Becht et al.4 argued that UMAP is preferable to t-SNE because it better preserves the global structure of the data and is more consistent across runs. Here we show that this alleged superiority of UMAP can be entirely attributed to different choices of initialization in the implementations used by Becht et al.: the t-SNE implementations by default used random initialization, while the UMAP implementation used a technique called Laplacian eigenmaps (LE)5 to initialize the embedding. We show that UMAP with random initialization preserves global structure as poorly as t-SNE with random initialization, while t-SNE with informative initialization performs as well as UMAP with informative initialization. On the basis of these observations, we argue that there is currently no evidence that the UMAP algorithm per se has any advantage over t-SNE in terms of preserving global structure. We also contend that these algorithms should always use informative initialization by default. At the core of both t-SNE and UMAP are loss functions that make similar points attract each other and push dissimilar points away from each other. Both algorithms minimize their loss functions by using gradient descent. Gradient descent begins with some initial configuration of points, and with each iteration the points are moved to decrease the loss function. The specific implementations of these algorithms used by Becht et al. differed in how the initial configuration of points was chosen: the t-SNE implementations placed the points randomly, whereas the UMAP implementation used LE5, an algorithm that can often achieve globally accurate embedding on its own6. The effect of this difference on how well the two algorithms preserve global structure was not discussed or investigated by Becht et al. We can illustrate the importance of initialization for both algorithms using a simple toy dataset (Fig. 1). We sampled n = 7,000 points from a circle with some added Gaussian noise and used UMAP and t-SNE to construct an embedding. We kept all parameters for both algorithms at their default values and only changed the initialization. The t-SNE algorithm with random initialization produced a knot; UMAP with random initialization produced a tangled web with multiple tears. In both cases, the global structure was evidently not preserved; indeed, gradient descent in t-SNE and UMAP only pulls close neighbors together and is not much influenced by the global arrangement of points. At the same time, LE recovers the original circle for this toy dataset and, when used for initialization, strongly improves the UMAP result. A method often recommended for initialization in t-SNE is principal component analysis (PCA)2. Here PCA also recovers the original circle and strongly improves the t-SNE result. In both cases, only with informative initialization can UMAP and t-SNE produce a faithful representation of the closed one-dimensional manifold—the circle. Using the code published by Becht et al., we analyzed the separate effects of initialization and algorithm on their results by adding UMAP with random initialization and t-SNE (using FIt-SNE7) with PCA initialization to the benchmarking comparison. Apart from the initialization, both algorithms were run with the same parameters as in Becht et al. We used all three datasets analyzed in the original publication (sample sizes from 320,000 to 820,000 cells)8–10. To quantify preservation of global structure, Becht et al. computed Pearson correlation between pairwise Euclidean distances in high-dimensional space and in the embedding. To quantify the reproducibility of the embedding, the authors embedded random subsamples of the data and measured the correlation of the coordinates of subsample embeddings with the coordinates of the full-dataset embeddings (up to symmetries around the coordinate axes). Our results show that t-SNE and UMAP with random initialization perform similarly poorly with regard to both metrics, whereas t-SNE and UMAP with PCA and LE initialization, respectively, perform similarly well (Table 1). See Extended Data Figs. 1–6 for the exact analogs of the original figures from Becht et al. Becht et al. wrote that their findings were “consistent with the idea” that UMAP performs “optimizations that are sensitive to global features of the data, thus reaching similar arrangements more consistently.” Our results show that this conclusion can be misleading: their findings were in fact not due to UMAP optimizations but rather to its initialization. We have recently argued that, for single-cell transcriptomic data, t-SNE with PCA initialization produces more meaningful embeddings than t-SNE with random initialization2. The findings of Becht et al., when interpreted correctly, also underscore the importance of using informative initialization and suggest that it should be used as the default option in t-SNE and UMAP implementations. PCA initialization has always been the default in openTSNE11, a Python reimplementation of FIt-SNE, and FIt-SNE v.1.2 now also uses it by default. OpenTSNE v.0.4 now also supports LE initialization. Importantly, t-SNE with non-random initialization should not be considered a new algorithm or even an extension of the original t-SNE; it is exactly the same algorithm with the same loss function, and almost any existing implementation trivially allows the use of any given initialization, including the PCA-based one. Initialization is critical for preserving global data structure in both t-SNE and UMAP
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
(2008) · Journal of Machine Learning Research · Cited by 33,051
Advances in neural information processing systems 7
Cios, K., & Shields, M. (1997) · Neurocomputing · doi:10.1016/s0925-2312(97)90018-7 · Cited by 22,059
UMAP: Uniform Manifold Approximation and Projection
McInnes, L., Healy, J., Saul, N., & Großberger, L. (2018) · Journal of Open Source Software · doi:10.21105/joss.00861 · Cited by 9,826
Dimensionality reduction for visualizing single-cell data using UMAP
Becht, E., McInnes, L., Healy, J., Dutertre, C.-A., Kwok, I. W. H., Ng, L. G., Ginhoux, F., & Newell, E. W. (2019) · Nature Biotechnology · doi:10.1038/nbt.4314 · Cited by 5,358
Laplacian Eigenmaps and Spectral Techniques for Embedding and Clustering
Belkin, M., & Niyogi, P. (2002) · Advances in Neural Information Processing Systems 14 · doi:10.7551/mitpress/1120.003.0080 · Cited by 4,412
Coifman, R. R., & Lafon, S. (2006) · Applied and Computational Harmonic Analysis · doi:10.1016/j.acha.2006.04.006 · Cited by 2,342
Mapping the Mouse Cell Atlas by Microwell-Seq
Han, X., Wang, R., Zhou, Y., Fei, L., Sun, H., Lai, S., Saadatpour, A., Zhou, Z., Chen, H., Ye, F., et al. (2018) · Cell · doi:10.1016/j.cell.2018.02.001 · Cited by 1,597
The art of using t-SNE for single-cell transcriptomics
Kobak, D., & Berens, P. (2019) · Nature Communications · doi:10.1038/s41467-019-13056-x · Cited by 1,115
Self-citationFUNDAMENTALS OF PLANARIAN REGENERATION
Reddien, P. W., & Alvarado, A. S. (2004) · Annual Review of Cell and Developmental Biology · doi:10.1146/annurev.cellbio.20.010403.095114 · Cited by 1,101
Fast interpolation-based t-SNE for improved visualization of single-cell RNA-seq data
Linderman, G. C., Rachh, M., Hoskins, J. G., Steinerberger, S., & Kluger, Y. (2019) · Nature Methods · doi:10.1038/s41592-018-0308-4 · Cited by 534
Self-citation
Cited by
UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
McInnes, L., Healy, J., & Melville, J. (2018) · arXiv · doi:10.48550/arxiv.1802.03426 · Cited by 7,101
CellRank for directed single-cell fate mapping
Lange, M., Bergen, V., Klein, M., Setty, M., Reuter, B., Bakhti, M., Lickert, H., Ansari, M., Schniering, J., Schiller, H. B., et al. (2022) · Nature Methods · doi:10.1038/s41592-021-01346-6 · Cited by 768
Toroidal topology of population activity in grid cells
Gardner, R. J., Hermansen, E., Pachitariu, M., Burak, Y., Baas, N. A., Dunn, B. A., Moser, M.-B., & Moser, E. I. (2022) · Nature · doi:10.1038/s41586-021-04268-7 · Cited by 428
Unsupervised Learning Methods for Molecular Simulation Data
Glielmo, A., Husic, B. E., Rodriguez, A., Clementi, C., Noé, F., & Laio, A. (2021) · Chemical Reviews · doi:10.1021/acs.chemrev.0c01195 · Cited by 402
The Specious Art of Single-Cell Genomics
Chari, T., & Pachter, L. (2021) · doi:10.1101/2021.08.25.457696 · Cited by 379
Metadata
- Authors
- Kobak, D., & Linderman, G. C.
- Venue
- Nature Biotechnology
- Published
- 2021-02
- Volume
- 39
- Issue
- 2
- Pages
- 156-157
- ISSN
- 0733-222X (ISSN-L)
- 1087-0156
- 1546-1696
- OpenAlex
- W3127452014
- Cited by
- 365
- References
- 15
- FWCI
- 15.50
- Type
- journal article
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Initialization is critical for preserving global data structure in both t-SNE and UMAP. https://app.cochat.ai/papers/10.1038/s41587-020-00809-z
- The paper
- Kobak, D., & Linderman, G. C. (2021). Initialization is critical for preserving global data structure in both t-SNE and UMAP. Nature Biotechnology. https://doi.org/10.1038/s41587-020-00809-z
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.