{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,20]],"date-time":"2026-03-20T00:17:18Z","timestamp":1773965838208,"version":"3.50.1"},"reference-count":33,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,1,2]],"date-time":"2026-01-02T00:00:00Z","timestamp":1767312000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Bioinform."],"abstract":"<jats:p>t-distributed Stochastic Neighbour Embedding (t-SNE) is a cornerstone for visualizing high-dimensional biological data, where each high-dimensional data point is represented as a point in a two-dimensional map. However, this static map provides no information about the stability of the visual layout, the features that influence it, or the impact of uncertainty in the input data. This work introduces a computational framework that allows one to extend the standard t-SNE plot by visual clues about the stability of the t-SNE embedding. First, we perform a sensitivity analysis to determine feature influence: by combining the Implicit Function Theorem with automatic differentiation, our method computes the sensitivity of the embedding w.r.t. the input data, provided in a Jacobian of first-order derivatives. Heatmap-visualizations of this Jacobian or summarizations thereof reveal which input features are most influential in shaping the embedding and identifying regions of structural instability. Second, when input data uncertainty is available, our framework uses this Jacobian to propagate error, probabilistically quantifying the positional uncertainty of each embedded point. This uncertainty is visualized by augmenting the plot with hypothetical outcomes, which display the positional confidence of each point. We apply our framework to three diverse biological datasets (bulk RNA-seq, proteomics, and single-cell transcriptomics), demonstrating its ability to directly link visual patterns to their underlying biological drivers and reveal ambiguities invisible in a standard plot. By providing this principled means to assess the robustness and interpretability of t-SNE visualizations, our work enables more rigorous and informed scientific conclusions in bioinformatics.<\/jats:p>","DOI":"10.3389\/fbinf.2025.1719516","type":"journal-article","created":{"date-parts":[[2026,1,2]],"date-time":"2026-01-02T06:28:14Z","timestamp":1767335294000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["Visualizing stability: a sensitivity analysis framework for t-SNE embeddings"],"prefix":"10.3389","volume":"5","author":[{"given":"Susanne","family":"Zabel","sequence":"first","affiliation":[]},{"given":"Philipp","family":"Hennig","sequence":"additional","affiliation":[]},{"given":"Kay","family":"Nieselt","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2026,1,2]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"12244","DOI":"10.1073\/pnas.1510227113","article-title":"Data-driven identification of prognostic tumor subpopulations using spatially mapped t-sne of mass spectrometry imaging data","volume":"113","author":"Abdelmoula","year":"2016","journal-title":"Proc. Natl. Acad. Sci."},{"key":"B2","doi-asserted-by":"publisher","first-page":"5415","DOI":"10.1038\/s41467-019-13055-y","article-title":"Automated optimized parameters for t-distributed stochastic neighbor embedding improve visualization and analysis of large datasets","volume":"10","author":"Belkina","year":"2019","journal-title":"Nat. Communications"},{"key":"B3","first-page":"393","article-title":"Explaining t-sne embeddings locally by adapting lime","author":"Bibal","year":"2020"},{"key":"B4","article-title":"JAX: composable transformations of Python+NumPy programs","author":"Bradbury","year":"2018"},{"key":"B5","first-page":"3","article-title":"R\u00e9sum\u00e9 d\u2019un m\u00e9moir sur la m\u00e9canique c\u00e9leste et sur un nouveau calcul appel\u00e9 calcul des limites","volume":"12","author":"Cauchy","year":"1831","journal-title":"Oeuvres Compl\u00e9tes d\u2019Augustun Cauchy"},{"key":"B6","first-page":"330","article-title":"t-sne-cuda: Gpu-accelerated t-sne and its applications to modern data","author":"Chan","year":"2018"},{"key":"B7","doi-asserted-by":"publisher","first-page":"154","DOI":"10.1177\/1473871615600010","article-title":"Explaining three-dimensional dimensionality reduction plots","volume":"15","author":"Coimbra","year":"2016","journal-title":"Inf. Vis."},{"key":"B8","doi-asserted-by":"publisher","first-page":"3690","DOI":"10.1007\/s10618-024-01055-6","article-title":"Gradient-based explanation for non-linear non-parametric dimensionality reduction","volume":"38","author":"Corbugy","year":"2024","journal-title":"Data Min. Knowl. Discov."},{"key":"B9","article-title":"Omnideconv: a unifying framework for using and benchmarking single-cell-informed deconvolution of bulk rna-seq data","author":"Dietrich","year":"2024","journal-title":"Biorxiv"},{"key":"B10","doi-asserted-by":"publisher","first-page":"2300061","DOI":"10.1002\/minf.202300061","article-title":"Feature importance-based interpretation of umap-visualized polymer space","volume":"42","author":"Ehiro","year":"2023","journal-title":"Mol. Inf."},{"key":"B11","first-page":"232","article-title":"A new supervised t-sne with dissimilarity measure for effective data visualization and classification","author":"Hajderanj","year":"2019"},{"key":"B12","doi-asserted-by":"publisher","first-page":"877","DOI":"10.1016\/j.neucom.2022.05.119","article-title":"Enhance explainability of manifold learning","volume":"500","author":"Han","year":"2022","journal-title":"Neurocomputing"},{"key":"B13","doi-asserted-by":"publisher","first-page":"3573","DOI":"10.1016\/j.cell.2021.04.048","article-title":"Integrated analysis of multimodal single-cell data","volume":"184","author":"Hao","year":"2021","journal-title":"Cell"},{"key":"B14","doi-asserted-by":"publisher","first-page":"417","DOI":"10.1037\/h0070888","article-title":"Analysis of a complex of statistical variables into principal components","volume":"24","author":"Hotelling","year":"1933","journal-title":"J. Educ. Psychol."},{"key":"B15","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s12859-021-04075-x","article-title":"Denoising large-scale biological data using network filters","volume":"22","author":"Kavran","year":"2021","journal-title":"BMC Bioinformatics"},{"key":"B16","doi-asserted-by":"publisher","first-page":"5416","DOI":"10.1038\/s41467-019-13056-x","article-title":"The art of using t-sne for single-cell transcriptomics","volume":"10","author":"Kobak","year":"2019","journal-title":"Nat. Communications"},{"key":"B17","volume-title":"The implicit function theorem: history, theory, and applications","author":"Krantz","year":"2002"},{"key":"B18","article-title":"Uncertainty-aware t-distributed stochastic neighbor embedding for single-cell rna-seq data","author":"Ma","year":"2024","journal-title":"arXiv Preprint arXiv:2410.00473"},{"key":"B19","article-title":"Umap: uniform manifold approximation and projection for dimension reduction","author":"McInnes","year":"2018","journal-title":"arXiv Preprint arXiv:1802.03426"},{"key":"B20","doi-asserted-by":"publisher","first-page":"164","DOI":"10.1109\/TVCG.2023.3326600","article-title":"Class-constrained t-sne: combining data features and class probabilities","volume":"30","author":"Meng","year":"2023","journal-title":"IEEE Trans. Vis. Comput. Graph."},{"key":"B21","doi-asserted-by":"publisher","first-page":"3981","DOI":"10.1007\/s13042-022-01635-2","article-title":"Confidence estimation for t-sne embeddings using random forest","volume":"13","author":"Ozgode Yigin","year":"2022","journal-title":"Int. J. Mach. Learn. Cybern."},{"key":"B22","doi-asserted-by":"publisher","first-page":"559","DOI":"10.1080\/14786440109462720","article-title":"Liii. on lines and planes of closest fit to systems of points in space","volume":"2","author":"Pearson","year":"1901","journal-title":"Lond. Edinb. Dublin philosophical Magazine Journal Science"},{"key":"B23","doi-asserted-by":"publisher","first-page":"e56883","DOI":"10.1371\/journal.pone.0056883","article-title":"Visualization of snps with t-sne","volume":"8","author":"Platzer","year":"2013","journal-title":"PloS One"},{"key":"B24","first-page":"1135","article-title":"\u201cWhy should i trust you?\u201d explaining the predictions of any classifier","author":"Ribeiro","year":"2016"},{"key":"B25","doi-asserted-by":"publisher","first-page":"2100103","DOI":"10.1002\/pmic.202100103","article-title":"A practical guide to interpreting and generating bottom-up proteomics data visualizations","volume":"22","author":"Schessner","year":"2022","journal-title":"Proteomics"},{"key":"B26","doi-asserted-by":"publisher","first-page":"101525","DOI":"10.1016\/j.isci.2020.101525","article-title":"Enzyme-constrained models and omics analysis of streptomyces coelicolor reveal metabolic changes that enhance heterologous production","volume":"23","author":"Sulheim","year":"2020","journal-title":"iScience"},{"key":"B27","first-page":"3221","article-title":"Accelerating t-sne using tree-based algorithms","volume":"15","author":"Van Der Maaten","year":"2014","journal-title":"J. Mach. Learn. Res."},{"key":"B28","article-title":"Visualizing data using t-sne","volume":"9","author":"Van der Maaten","year":"2008","journal-title":"J. Mach. Learn. Res."},{"key":"B29","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1186\/s13059-017-1382-0","article-title":"Scanpy: large-scale single-cell gene expression data analysis","volume":"19","author":"Wolf","year":"2018","journal-title":"Genome Biology"},{"key":"B30","doi-asserted-by":"publisher","first-page":"2011","DOI":"10.1109\/TVCG.2023.3345532","article-title":"VIPurPCA: visualizing and propagating uncertainty in principal component analysis","volume":"30","author":"Zabel","year":"2023","journal-title":"IEEE Trans. Vis. Comput. Graph."},{"key":"B31","doi-asserted-by":"publisher","first-page":"109105","DOI":"10.1016\/j.compbiomed.2024.109105","article-title":"Mesoscopic structure graphs for interpreting uncertainty in non-linear embeddings","volume":"182","author":"Zhao","year":"2024","journal-title":"Comput. Biol. Med."},{"key":"B32","doi-asserted-by":"publisher","first-page":"14049","DOI":"10.1038\/ncomms14049","article-title":"Massively parallel digital transcriptional profiling of single cells","volume":"8","author":"Zheng","year":"2017","journal-title":"Nat. Communications"},{"key":"B33","first-page":"159","article-title":"Visualization of single cell rna-seq data using t-sne in r","author":"Zhou","year":"2020"}],"container-title":["Frontiers in Bioinformatics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fbinf.2025.1719516\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,2]],"date-time":"2026-01-02T06:28:16Z","timestamp":1767335296000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fbinf.2025.1719516\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,2]]},"references-count":33,"alternative-id":["10.3389\/fbinf.2025.1719516"],"URL":"https:\/\/doi.org\/10.3389\/fbinf.2025.1719516","relation":{},"ISSN":["2673-7647"],"issn-type":[{"value":"2673-7647","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1,2]]},"article-number":"1719516"}}