IDEAS home Printed from https://ideas.repec.org/a/nat/natcom/v16y2025i1d10.1038_s41467-025-60434-9.html
   My bibliography  Save this article

Assessing and improving reliability of neighbor embedding methods: a map-continuity perspective

Author

Listed:
  • Zhexuan Liu

    (University of Wisconsin-Madison)

  • Rong Ma

    (Harvard University
    Dana-Farber Cancer Institute)

  • Yiqiao Zhong

    (University of Wisconsin-Madison)

Abstract

Visualizing high-dimensional data is essential for understanding biomedical data and deep learning models. Neighbor embedding methods, such as t-SNE and UMAP, are widely used but can introduce misleading visual artifacts. We find that the manifold learning interpretations from many prior works are inaccurate and that the misuse stems from a lack of data-independent notions of embedding maps, which project high-dimensional data into a lower-dimensional space. Leveraging the leave-one-out principle, we introduce LOO-map, a framework that extends embedding maps beyond discrete points to the entire input space. We identify two forms of map discontinuity that distort visualizations: one exaggerates cluster separation and the other creates spurious local structures. As a remedy, we develop two types of point-wise diagnostic scores to detect unreliable embedding points and improve hyperparameter selection, which are validated on datasets from computer vision and single-cell omics.

Suggested Citation

  • Zhexuan Liu & Rong Ma & Yiqiao Zhong, 2025. "Assessing and improving reliability of neighbor embedding methods: a map-continuity perspective," Nature Communications, Nature, vol. 16(1), pages 1-16, December.
  • Handle: RePEc:nat:natcom:v:16:y:2025:i:1:d:10.1038_s41467-025-60434-9
    DOI: 10.1038/s41467-025-60434-9
    as

    Download full text from publisher

    File URL: https://www.nature.com/articles/s41467-025-60434-9
    File Function: Abstract
    Download Restriction: no

    File URL: https://libkey.io/10.1038/s41467-025-60434-9?utm_source=ideas
    LibKey link: if access is restricted and if your library uses this service, LibKey will redirect you to where you can use your library subscription to access this item
    ---><---

    References listed on IDEAS

    as
    1. Md Tauhidul Islam & Zixia Zhou & Hongyi Ren & Masoud Badiei Khuzani & Daniel Kapp & James Zou & Lu Tian & Joseph C. Liao & Lei Xing, 2023. "Revealing hidden patterns in deep neural network feature space continuum via manifold learning," Nature Communications, Nature, vol. 14(1), pages 1-20, December.
    2. Dmitry Kobak & Philipp Berens, 2019. "The art of using t-SNE for single-cell transcriptomics," Nature Communications, Nature, vol. 10(1), pages 1-14, December.
    3. Karsten Bach & Sara Pensa & Marta Grzelak & James Hadfield & David J. Adams & John C. Marioni & Walid T. Khaled, 2017. "Differentiation dynamics of mammary epithelial cells revealed by single-cell RNA sequencing," Nature Communications, Nature, vol. 8(1), pages 1-11, December.
    4. Tara Chari & Lior Pachter, 2023. "The specious art of single-cell genomics," PLOS Computational Biology, Public Library of Science, vol. 19(8), pages 1-20, August.
    5. Lucy Xia & Christy Lee & Jingyi Jessica Li, 2024. "Statistical method scDEED for detecting dubious 2D single-cell embeddings and optimizing t-SNE and UMAP hyperparameters," Nature Communications, Nature, vol. 15(1), pages 1-21, December.
    6. Anna C. Belkina & Christopher O. Ciccolella & Rina Anno & Richard Halpert & Josef Spidlen & Jennifer E. Snyder-Cappione, 2019. "Automated optimized parameters for T-distributed stochastic neighbor embedding improve visualization and analysis of large datasets," Nature Communications, Nature, vol. 10(1), pages 1-12, December.
    7. Tetsutaro Hayashi & Haruka Ozaki & Yohei Sasagawa & Mana Umeda & Hiroki Danno & Itoshi Nikaido, 2018. "Single-cell full-length total RNA sequencing uncovers dynamics of recursive splicing and enhancer RNAs," Nature Communications, Nature, vol. 9(1), pages 1-16, December.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Rong Ma & Eric D. Sun & James Zou, 2023. "A spectral method for assessing and combining multiple data visualizations," Nature Communications, Nature, vol. 14(1), pages 1-14, December.
    2. Lucy Xia & Christy Lee & Jingyi Jessica Li, 2024. "Statistical method scDEED for detecting dubious 2D single-cell embeddings and optimizing t-SNE and UMAP hyperparameters," Nature Communications, Nature, vol. 15(1), pages 1-21, December.
    3. Chen Jiang & Alessia Centonze & Yura Song & Antonius Chrisnandy & Elisavet Tika & Saba Rezakhani & Zahra Zahedi & Gaëlle Bouvencourt & Christine Dubois & Alexandra Van Keymeulen & Matthias Lütolf & Al, 2024. "Collagen signaling and matrix stiffness regulate multipotency in glandular epithelial stem cells in mice," Nature Communications, Nature, vol. 15(1), pages 1-19, December.
    4. Kai Battenberg & S. Thomas Kelly & Radu Abu Ras & Nicola A. Hetherington & Makoto Hayashi & Aki Minoda, 2022. "A flexible cross-platform single-cell data processing pipeline," Nature Communications, Nature, vol. 13(1), pages 1-7, December.
    5. Hinako M Takase & Tappei Mishina & Tetsutaro Hayashi & Mika Yoshimura & Mariko Kuse & Itoshi Nikaido & Tomoya S Kitajima, 2024. "Transcriptomic signatures of WNT-driven pathways and granulosa cell-oocyte interactions during primordial follicle activation," PLOS ONE, Public Library of Science, vol. 19(10), pages 1-26, October.
    6. Guocheng Fang & Zhen Qiao & Luqi Huang & Hui Zhu & Jun Xie & Tian Zhou & Zhongshu Xiong & I-Hsin Su & Dayong Jin & Yu-Cheng Chen, 2024. "Single-cell laser emitting cytometry for label-free nucleolus fingerprinting," Nature Communications, Nature, vol. 15(1), pages 1-13, December.
    7. Ana Sofia Rocha & Alejandro Collado-Solé & Osvaldo Graña-Castro & Jaime Redondo-Pedraza & Gonzalo Soria-Alcaide & Alex Cordero & Patricia G. Santamaría & Eva González-Suárez, 2023. "Luminal Rank loss decreases cell fitness leading to basal cell bipotency in parous mammary glands," Nature Communications, Nature, vol. 14(1), pages 1-13, December.
    8. Kaiwen Wang & Yuqiu Yang & Fangjiang Wu & Bing Song & Xinlei Wang & Tao Wang, 2023. "Comparative analysis of dimension reduction methods for cytometry by time-of-flight data," Nature Communications, Nature, vol. 14(1), pages 1-18, December.
    9. Wanying Wu & Jinyang Zhang & Xiaofei Cao & Zhengyi Cai & Fangqing Zhao, 2022. "Exploring the cellular landscape of circular RNAs using full-length single-cell RNA sequencing," Nature Communications, Nature, vol. 13(1), pages 1-14, December.
    10. Renigier-Biłozor Małgorzata & Janowski Artur, 2024. "Human-Machine Synergy in Real Estate Similarity Concept," Real Estate Management and Valuation, Sciendo, vol. 32(2), pages 13-30.
    11. L. Mathur & B. Szalai & N. H. Du & R. Utharala & M. Ballinger & J. J. M. Landry & M. Ryckelynck & V. Benes & J. Saez-Rodriguez & C. A. Merten, 2022. "Combi-seq for multiplexed transcriptome-based profiling of drug combinations using deterministic barcoding in single-cell droplets," Nature Communications, Nature, vol. 13(1), pages 1-15, December.
    12. Boumediene Ladjal & Imad Eddine Tibermacine & Mohcene Bechouat & Moussa Sedraoui & Christian Napoli & Abdelaziz Rabehi & Djemoui Lalmi, 2024. "Hybrid models for direct normal irradiance forecasting: a case study of Ghardaia zone (Algeria)," Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards, Springer;International Society for the Prevention and Mitigation of Natural Hazards, vol. 120(15), pages 14703-14725, December.
    13. Nils-Petter Rudqvist & Maud Charpentier & Claire Lhuillier & Erik Wennerberg & Sheila Spada & Caroline Sheridan & Xi Kathy Zhou & Tuo Zhang & Silvia C. Formenti & Jennifer S. Sims & Alicia Alonso & Sa, 2023. "Immunotherapy targeting different immune compartments in combination with radiation therapy induces regression of resistant tumors," Nature Communications, Nature, vol. 14(1), pages 1-23, December.
    14. Andriana Manousidaki & Anna Little & Yuying Xie, 2024. "Clustering and visualization of single-cell RNA-seq data using path metrics," PLOS Computational Biology, Public Library of Science, vol. 20(5), pages 1-19, May.
    15. Yanyan Diao & Dandan Liu & Huan Ge & Rongrong Zhang & Kexin Jiang & Runhui Bao & Xiaoqian Zhu & Hongjie Bi & Wenjie Liao & Ziqi Chen & Kai Zhang & Rui Wang & Lili Zhu & Zhenjiang Zhao & Qiaoyu Hu & Ho, 2023. "Macrocyclization of linear molecules by deep learning to facilitate macrocyclic drug candidates discovery," Nature Communications, Nature, vol. 14(1), pages 1-15, December.
    16. Adrià Fernández-Torras & Miquel Duran-Frigola & Martino Bertoni & Martina Locatelli & Patrick Aloy, 2022. "Integrating and formatting biomedical data as pre-calculated knowledge graph embeddings in the Bioteque," Nature Communications, Nature, vol. 13(1), pages 1-18, December.
    17. Mohammad Mosaffa & Omid Rafieian & Hema Yoganarasimhan, 2025. "Visual Polarization Measurement Using Counterfactual Image Generation," Papers 2503.10738, arXiv.org.
    18. Maryam Ghaderi Najafabadi & G. Kenneth Gray & Li Ren Kong & Komal Gupta & David Perera & Huw Naylor & Joan S. Brugge & Ashok R. Venkitaraman & Mona Shehata, 2023. "A transcriptional response to replication stress selectively expands a subset of Brca2-mutant mammary epithelial cells," Nature Communications, Nature, vol. 14(1), pages 1-17, December.
    19. Zhiyuan Yuan & Yisi Li & Minglei Shi & Fan Yang & Juntao Gao & Jianhua Yao & Michael Q. Zhang, 2022. "SOTIP is a versatile method for microenvironment modeling with spatial omics data," Nature Communications, Nature, vol. 13(1), pages 1-19, December.
    20. Naziia Kurmasheva & Aida Said & Boaz Wong & Priscilla Kinderman & Xiaoying Han & Anna H. F. Rahimic & Alena Kress & Madalina E. Carter-Timofte & Emilia Holm & Demi Horst & Christoph F. Kollmann & Zhen, 2024. "Octyl itaconate enhances VSVΔ51 oncolytic virotherapy by multitarget inhibition of antiviral and inflammatory pathways," Nature Communications, Nature, vol. 15(1), pages 1-28, December.

    More about this item

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:nat:natcom:v:16:y:2025:i:1:d:10.1038_s41467-025-60434-9. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Sonal Shukla or Springer Nature Abstracting and Indexing (email available below). General contact details of provider: http://www.nature.com .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.