CoChat
Sign Up
Cited by 20Single-cell and spatial transcriptomicsGene Regulatory Network AnalysisGene expression and cancer classification

PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis

Open access
Wu, Yan, Wershof, Esther, Schmon, Sebastian, Nassar, Marcel, Osiński, Błażej, Eksi, Ridvan, Yan, Zichao, Stark, Rory +2 more·2025·Advances in Neural Information Processing Systems 38
DOI 10.52202/085713-3225
Open this paper in CoChat — free accountOpen PDF

Abstract

We introduce a comprehensive framework for modeling single cell transcriptomic responses to perturbations, aimed at standardizing benchmarking in this rapidly evolving field. Our approach includes a modular and user-friendly model development and evaluation platform, a collection of diverse perturbational datasets, and a set of metrics designed to fairly compare models and dissect their performance. Through extensive evaluation of both published and baseline models across diverse datasets, we highlight the limitations of widely used models, such as mode collapse. We also demonstrate the importance of rank metrics which complement traditional model fit measures, such as RMSE, for validating model effectiveness. Notably, our results show that while no single model architecture clearly outperforms others, simpler architectures are generally competitive and scale well with larger datasets. Overall, this benchmarking exercise sets new standards for model evaluation, supports robust model development, and furthers the use of these models to simulate genetic and chemical screens for therapeutic discovery.

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.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.52202/085713-3225

Cites

Statistics

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.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • The evaluation of prior benchmarks was limited in dataset diversity and task coverage, leaving open whether their conclusions generalize across broader biological contexts.

    “While this work provides valuable insights, its evaluation was limited in dataset diversity and task coverage, leaving the question of whether such conclusions generalize across broader biological contexts.”

    From A systematic comparison of single-cell perturbation response prediction models (2026) · cites this paper as [23]

    PerturBench describes its own evaluation in the Abstract as extensive and conducted across diverse datasets. The ‘Biologically relevant tasks’ section lists six datasets and several task types, including covariate transfer, combination prediction, data scaling, and imbalanced-data scenarios. That is contrary to the claim that this benchmark’s evaluation was limited in dataset diversity and task coverage; the paper does not claim that its findings generalize to every biological context. (LLM, conf=0.78)

    Checked by a reasoning model — may be wrong.

    Experimental — accuracy not yet measured

  • Generative models often fail to outperform methods that do not explicitly model distributional structure when applied to the prediction of perturbation effects.

    “However, in practice, these models often fail to outperform methods that do not explicitly model distributional structure when applied to the prediction of perturbation effects.”

    From Predicting cellular responses to perturbation across diverse contexts with State (2026) · cites this paper as [15]

    The Discussion reports that, for predicting average perturbation effects, no model class is clearly superior and that simpler matching-based models outperform CPA* and SAMS-VAE* on larger datasets. But it also reports that those autoencoder models outperform the baselines on distributional metrics. The claim omits this distinction between average-effect and distributional prediction. (LLM, conf=0.86)

    Experimental — accuracy not yet measured

  • Despite the rapid growth of perturbation datasets in size and scope, proportional gains in predictive capabilities have not been achieved.

    “Despite the rapid growth of perturbation datasets in size and scope, proportional gains in predictive capabilities have not been achieved.”

    From Predicting cellular responses to perturbation across diverse contexts with State (2026) · cites this paper as [15]

    The Discussion (section 6) reports that models perform worse on larger datasets and tasks, supporting the concern about predictive limitations. But the data-scaling results (section 5.3) say most models improve with more training data. The paper does not establish that gains in predictive capability have failed to keep pace with the growth of perturbation datasets in size and scope. (LLM, conf=0.78)

    Experimental — accuracy not yet measured

  • Current deep learning methods do not consistently outperform linear models when generalizing perturbation effects across cellular contexts.

    “Current deep learning methods do not consistently outperform linear models when generalizing perturbation effects across cellular contexts.”

    From Predicting cellular responses to perturbation across diverse contexts with State (2026) · cites this paper as [15]

    Section 6 (Discussion; page unavailable) says no model or class is clearly best at predicting average perturbation effects, and §5.1 reports that simpler baselines sometimes outperform more complex models across cell lines. But the paper distinguishes those baselines from its linear model; these passages do not establish that deep learning methods fail to consistently outperform linear models specifically when generalizing across cellular contexts. (LLM, conf=0.82)

    Experimental — accuracy not yet measured

  • Recent studies reported that complex models do not consistently outperform simpler alternatives for perturbation effect prediction.

    “Recent studies have provided initial benchmarks comparing foundation models, other deep learning architectures, and simple baselines for perturbation effect prediction and reported that complex models do not consistently outperform the simpler alternatives (22–26).”

    From A systematic comparison of single-cell perturbation response prediction models (2026) · cites this paper as [23]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

No references could be resolved yet.

Research works printed but not linked: 53

  1. Adamson, B., Norman, T. M., Jost, M., Cho, M. Y., Nuñez, J. K., Chen, Y., Villalta, J. E., Gilbert, L. A., Horlbeck, M. A., Hein, M. Y., et al. (2016). A multiplexed single-cell crispr screening platform enables systematic dissection of the unfolded protein response. Cell, 167(7):1867-1882.
  2. Adduri, A. K., Gautam, D., Bevilacqua, B., Imran, A., Shah, R., Naghipourfar, M., Teyssier, N., Ilango, R., Nagaraj, S., Dong, M., Ricci-Tam, C., Carpenter, C., Subramanyam, V., Winters, A., Tirukkovular, S., Sullivan, J., Plosky, B. S., Eraslan, B., Youngblut, N. D., Leskovec, J., Gilbert, L. A., Konermann, S., Hsu, P. D., Dobin, A., Burke, D. P., Goodarzi, H., and Roohani, Y. H. (2025). Predicting cellular responses to perturbation across diverse contexts with state. bioRxiv.
  3. Ahlmann-Eltze, C., Huber, W., and Anders, S. (2024). Deep learning-based predictions of gene perturbation effects do not yet outperform simple linear methods. bioRxiv, page 2024.09.16.613342.
  4. Akiba, T., Sano, S., Yanase, T., Ohta, T., and Koyama, M. (2019). Optuna: A next-generation hyperparameter optimization framework. In Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining, pages 2623-2631.
  5. Ba, J. L., Kiros, J. R., and Hinton, G. E. (2016). Layer normalization. arXiv preprint arXiv:1607.06450. · arXiv
  6. Bengio, Y., Courville, A., and Vincent, P. (2013). Representation learning: A review and new perspectives. IEEE transactions on pattern analysis and machine intelligence, 35(8):1798-1828.
  7. Bereket, M. and Karaletsos, T. (2024). Modelling cellular perturbations with the sparse additive mechanism shift variational autoencoder. Advances in Neural Information Processing Systems, 36.
  8. Bergstra, J., Bardenet, R., Bengio, Y., and Kégl, B. (2011). Algorithms for hyper-parameter optimization. Advances in neural information processing systems, 24.
  9. Bock, C., Datlinger, P., Chardon, F., Coelho, M. A., Dong, M. B., Lawson, K. A., Lu, T., Maroc, L., Norman, T. M., Song, B., Stanley, G., Chen, S., Garnett, M., Li, W., Moffat, J., Qi, L. S., Shapiro, R. S., Shendure, J., Weissman, J. S., and Zhuang, X. (2022). High-content CRISPR screening. Nat. Rev. Methods Primers, 2(8):1-23.
  10. Bunne, C., Stark, S. G., Gut, G., del Castillo, J. S., Levesque, M., Lehmann, K.-V., Pelkmans, L., Krause, A., and Rätsch, G. (2023). Learning single-cell perturbation responses using neural optimal transport. Nat. Methods, 20(11):1759-1768.
  11. Chandrasekaran, S. N., Ackerman, J., Alix, E., Ando, D. M., Arevalo, J., Bennion, M., Boisseau, N., Borowa, A., Boyd, J. D., Brino, L., Byrne, P. J., Ceulemans, H., Ch'ng, C., Cimini, B. A., Clevert, D.-A., Deflaux, N., Doench, J. G., Dorval, T., Doyonnas, R., Dragone, V., Engkvist, O., Faloon, P. W., Fritchman, B., Fuchs, F., Garg, S., Gilbert, T. J., Glazer, D., Gnutt, D., Goodale, A., Grignard, J., Guenther, J., Han, Y., Hanifehlou, Z., Hariharan, S., Hernandez, D., Horman, S. R., Hormel, G., Huntley, M., Icke, I., Iida, M., Jacob, C. B., Jaensch, S., Khetan, J., Kost-Alimova, M., Krawiec, T., Kuhn, D., Lardeau, C.-H., Lembke, A., Lin, F., Little, K. D., Lofstrom, K. R., Lotfi, S., Logan, D. J., Luo, Y., Madoux, F., Marin Zapata, P. A., Marion, B. A., Martin, G., McCarthy, N. J., Mervin, L., Miller, L., Mohamed, H., Monteverde, T., Mouchet, E., Nicke, B., Ogier, A., Ong, A.-L., Osterland, M., Otrocka, M., Peeters, P. J., Pilling, J., Prechtl, S., Qian, C., Rataj, K., Root, D. E., Sakata, S. K., Scrace, S., Shimizu, H., Simon, D., Sommer, P., Spruiell, C., Sumia, I., Swalley, S. E., Terauchi, H., Thibaudeau, A., Unruh, A., Van de Waeter, J., Van Dyck, M., van Staden, C., Warchoł, M., Weisbart, E., Weiss, A., Wiest-Daessle, N., Williams, G., Yu, S., Zapiec, B., Żyła, M., Singh, S., and Carpenter, A. E. (2023). Jump cell painting dataset: morphological impact of 136, 000 chemical and genetic perturbations. bioRxiv.
  12. Cui, H., Wang, C., Maan, H., Pang, K., Luo, F., Duan, N., and Wang, B. (2024). scGPT: toward building a foundation model for single-cell multi-omics using generative AI. Nat. Methods, pages 1-11.
  13. Cui, Y., Jia, M., Lin, T.-Y., Song, Y., and Belongie, S. (2019). Class-balanced loss based on effective number of samples. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 9268-9277.
  14. Dixit, A., Parnas, O., Li, B., Chen, J., Fulco, C. P., Jerby-Arnon, L., Marjanovic, N. D., Dionne, D., Burks, T., Raychowdhury, R., et al. (2016). Perturb-seq: dissecting molecular circuits with scalable single-cell rna profiling of pooled genetic screens. cell, 167(7):1853-1866.
  15. Frangieh, C. J., Melms, J. C., Thakore, P. I., Geiger-Schuller, K. R., Ho, P., Luoma, A. M., Cleary, B., Jerby-Arnon, L., Malu, S., Cuoco, M. S., Zhao, M., Ager, C. R., Rogava, M., Hovey, L., Rotem, A., Bernatchez, C., Wucherpfennig, K. W., Johnson, B. E., Rozenblatt-Rosen, O., Schadendorf, D., Regev, A., and Izar, B. (2021). Multimodal pooled Perturb-CITE-seq screens in patient models define mechanisms of cancer immune evasion. Nat. Genet., 53(3):332-341.
  16. Gavriilidis, G. I., Vasileiou, V., Orfanou, A., Ishaque, N., and Psomopoulos, F. (2024). A mini- review on perturbation modelling across single-cell omic modalities. Comput. Struct. Biotechnol. J., 23:1886.
  17. Hao, M., Gong, J., Zeng, X., Liu, C., Guo, Y., Cheng, X., Wang, T., Ma, J., Song, L., and Zhang, X. (2023). Large Scale Foundation Model on Single-cell Transcriptomics. bioRxiv, page 2023.05.29.542705.
  18. Jaeger, S., Fulle, S., and Turk, S. (2018). Mol2vec: Unsupervised Machine Learning Approach with Chemical Intuition. J. Chem. Inf. Model., 58(1):27-35.
  19. Ji, Y., Green, T., Peidli, S., Bahrami, M., Liu, M., Zappia, L., Hrovatin, K., Sander, C., and Theis, F. (2023). Optimal distance metrics for single-cell rna-seq populations. bioRxiv, pages 2023-12.
  20. Jiang, L., Dalgarno, C., Papalexi, E., Mascio, I., Wessels, H.-H., Yun, H., Iremadze, N., Lithwick- Yanai, G., Lipson, D., and Satija, R. (2024a). Systematic reconstruction of molecular pathway signatures using scalable single-cell perturbation screens. bioRxiv, page 2024.01.29.576933.
  21. Kaplan, J., McCandlish, S., Henighan, T., Brown, T. B., Chess, B., Child, R., Gray, S., Radford, A., Wu, J., and Amodei, D. (2020). Scaling laws for neural language models. arXiv preprint arXiv:2001.08361. · arXiv
  22. Keenan, A. B., Jenkins, S. L., Jagodnik, K. M., Koplev, S., He, E., Torre, D., Wang, Z., Dohlman, A. B., Silverstein, M. C., Lachmann, A., Kuleshov, M. V., Ma'ayan, A., Stathias, V., Terryn, R., Cooper, D., Forlin, M., Koleti, A., Vidovic, D., Chung, C., Schürer, S. C., Vasiliauskas, J., Pilarczyk, M., Shamsaei, B., Fazel, M., Ren, Y., Niu, W., Clark, N. A., White, S., Mahi, N., Zhang, L., Kouril, M., Reichard, J. F., Sivaganesan, S., Medvedovic, M., Meller, J., Koch, R. J., Birtwistle, M. R., Iyengar, R., Sobie, E. A., Azeloglu, E. U., Kaye, J., Osterloh, J., Haston, K., Kalra, J., Finkbiener, S., Li, J., Milani, P., Adam, M., Escalante-Chong, R., Sachs, K., Lenail, A., Ramamoorthy, D., Fraenkel, E., Daigle, G., Hussain, U., Coye, A., Rothstein, J., Sareen, D., Ornelas, L., Banuelos, M., Mandefro, B., Ho, R., Svendsen, C. N., Lim, R. G., Stocksdale, J., Casale, M. S., Thompson, T. G., Wu, J., Thompson, L. M., Dardov, V., Venkatraman, V., Matlock, A., Van Eyk, J. E., Jaffe, J. D., Papanastasiou, M., Subramanian, A., Golub, T. R., Erickson, S. D., Fallahi-Sichani, M., Hafner, M., Gray, N. S., Lin, J.-R., Mills, C. E., Muhlich, J. L., Niepel, M., Shamu, C. E., Williams, E. H., Wrobel, D., Sorger, P. K., Heiser, L. M., Gray, J. W., Korkola, J. E., Mills, G. B., LaBarge, M., Feiler, H. S., Dane, M. A., Bucher, E., Nederlof, M., Sudar, D., Gross, S., Kilburn, D. F., Smith, R., Devlin, K., Margolis, R., Derr, L., Lee, A., and Pillai, A. (2018). The library of integrated network-based cellular signatures nih program: System-level cataloging of human cells response to perturbations. Cell Systems, 6(1):13-24.
  23. Kernfeld, E., Yang, Y., Weinstock, J. S., Battle, A., and Cahan, P. (2023). A systematic comparison of computational methods for expression forecasting. bioRxiv.
  24. Klein, D., Fleck, J. S., Bobrovskiy, D., Zimmermann, L., Becker, S., Palma, A., Dony, L., Tejada-Lapuerta, A., Huguet, G., Lin, H.-C., Azbukina, N., Sanchís-Calleja, F., Uscidda, T., Szalata, A., Gander, M., Regev, A., Treutlein, B., Camp, J. G., and Theis, F. J. (2025). Cellflow enables generative single-cell phenotype modeling with flow matching. bioRxiv.
  25. Lachapelle, S., Rodriguez, P., Sharma, Y., Everett, K. E., Le Priol, R., Lacoste, A., and Lacoste- Julien, S. (2022). Disentanglement via mechanism sparsity regularization: A new principle for nonlinear ica. In Conference on Causal Learning and Reasoning, pages 428-484. PMLR.
  26. Li, C., Gao, H., She, Y., Bian, H., Chen, Q., Liu, K., Wei, L., and Zhang, X. (2024a). Bench- marking ai models forin silicogene perturbation of cells. bioRxiv.
  27. Li, L., You, Y., Liao, W., Fan, X., Lu, S., Cao, Y., Li, B., Ren, W., Fu, Y., Kong, J., Zheng, S., Chen, J., Liu, X., and Tian, L. (2024b). A systematic comparison of single-cell perturbation response prediction models. bioRxiv.
  28. Lopez, R., Tagasovska, N., Ra, S., Cho, K., Pritchard, J. K., and Regev, A. (2022). Learning Causal Representations of Single Cells via Sparse Mechanism Shift Modeling. arXiv.
  29. Lowe, R., Shirley, N., Bleackley, M., Dolan, S., and Shafee, T. (2017). Transcriptomics technologies. PLoS computational biology, 13(5):e1005457.
  30. Macosko, E. Z., Basu, A., Satija, R., Nemesh, J., Shekhar, K., Goldman, M., Tirosh, I., Bialas, A. R., Kamitaki, N., Martersteck, E. M., Trombetta, J. J., Weitz, D. A., Sanes, J. R., Shalek, A. K., Regev, A., and McCarroll, S. A. (2015). Highly Parallel Genome-wide Expression Profiling of Individual Cells Using Nanoliter Droplets. Cell, 161(5):1202-1214.
  31. McFaline-Figueroa, J. L., Srivatsan, S., Hill, A. J., Gasperini, M., Jackson, D. L., Saunders, L., Domcke, S., Regalado, S. G., Lazarchuck, P., Alvarez, S., et al. (2024). Multiplex single- cell chemical genomics reveals the kinase dependence of the response to targeted therapy. Cell Genomics, 4(2).
  32. Norman, T. M., Horlbeck, M. A., Replogle, J. M., Ge, A. Y., Xu, A., Jost, M., Gilbert, L. A., and Weissman, J. S. (2019). Exploring genetic interaction manifolds constructed from rich single-cell phenotypes. Science, 365(6455):786-793.
  33. Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., Desmaison, A., Köpf, A., Yang, E. Z., DeVito, Z., Raison, M., Tejani, A., Chilamkurthy, S., Steiner, B., Fang, L., Bai, J., and Chintala, S. (2019). Pytorch: An imperative style, high-performance deep learning library. Advances in Neural Information Processing Systems, abs/1912.01703. · arXiv
  34. Peidli, S., Green, T. D., Shen, C., Gross, T., Min, J., Garda, S., Yuan, B., Schumacher, L. J., Taylor-King, J. P., Marks, D. S., Luna, A., Blüthgen, N., and Sander, C. (2024). scPerturb: harmonized single-cell perturbation data. Nat. Methods, 21(3):531-540.
  35. Piran, Z., Cohen, N., Hoshen, Y., and Nitzan, M. (2024). Disentanglement of single-cell data with biolord. Nat. Biotechnol., pages 1-6.
  36. Radev, D. R., Qi, H., Wu, H., and Fan, W. (2002). Evaluating web-based question answering systems. In González Rodríguez, M. and Suarez Araujo, C. P., editors, Proceedings of the Third International Conference on Language Resources and Evaluation (LREC'02), Las Palmas, Canary Islands -Spain. European Language Resources Association (ELRA).
  37. Replogle, J. M., Saunders, R. A., Pogson, A. N., Hussmann, J. A., Lenail, A., Guna, A., Masci- broda, L., Wagner, E. J., Adelman, K., Lithwick-Yanai, G., Iremadze, N., Oberstrass, F., Lipson, D., Bonnar, J. L., Jost, M., Norman, T. M., and Weissman, J. S. (2022). Mapping information-rich genotype-phenotype landscapes with genome-scale Perturb-seq. Cell, 185(14):2559-2575.e28.
  38. Reymond, J.-L. (2015). The Chemical Space Project. Acc. Chem. Res., 48(3):722-730.
  39. Rives, A., Meier, J., Sercu, T., Goyal, S., Lin, Z., Liu, J., Guo, D., Ott, M., Zitnick, C. L., Ma, J., and Fergus, R. (2021). Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences. Proc. Natl. Acad. Sci. U.S.A., 118(15):e2016239118.
  40. Roohani, Y., Huang, K., and Leskovec, J. (2023). Predicting transcriptional outcomes of novel multigene perturbations with GEARS. Nat. Biotechnol., pages 1-9.
  41. Roohani, Y. H., Hua, T. J., Tung, P.-Y., Bounds, L. R., Yu, F. B., Dobin, A., Teyssier, N., Adduri, A., Woodrow, A., Plosky, B. S., Mehta, R., Hsu, B., Sullivan, J., Ricci-Tam, C., Li, N., Kazaks, J., Gilbert, L. A., Konermann, S., Hsu, P. D., Goodarzi, H., and Burke, D. P. (2025). Virtual cell challenge: Toward a turing test for the virtual cell. Cell, 188(13):3370-3374.
  42. Ross, J., Belgodere, B., Chenthamarakshan, V., Padhi, I., Mroueh, Y., and Das, P. (2022). Large-scale chemical language representations capture molecular structure and properties. Nat. Mach. Intell., 4(12):1256-1264.
  43. Shalem, O., Sanjana, N. E., Hartenian, E., Shi, X., Scott, D. A., Mikkelsen, T. S., Heckl, D., Ebert, B. L., Root, D. E., Doench, J. G., and Zhang, F. (2014). Genome-Scale CRISPR-Cas9 Knockout Screening in Human Cells. Science, 343(6166):84-87.
  44. Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., and Salakhutdinov, R. (2014). Dropout: a simple way to prevent neural networks from overfitting. The journal of machine learning research, 15(1):1929-1958.
  45. Srivatsan, S. R., McFaline-Figueroa, J. L., Ramani, V., Saunders, L., Cao, J., Packer, J., Pliner, H. A., Jackson, D. L., Daza, R. M., Christiansen, L., Zhang, F., Steemers, F., Shendure, J., and Trapnell, C. (2020). Massively multiplex chemical transcriptomics at single-cell resolution. Science, 367(6473):45-51.
  46. Stuart, E. A. (2010). Matching methods for causal inference: A review and a look forward. Statistical science: a review journal of the Institute of Mathematical Statistics, 25(1):1.
  47. Szałata, A., Benz, A., Cannoodt, R., Cortes, M., Fong, J., Kuppasani, S., Lieberman, R., Liu, T., Mas-Rosario, J. A., Meinl, R., et al. (2024). A benchmark for prediction of transcriptomic responses to chemical perturbations across cell types. Advances in Neural Information Processing Systems, 37:20566-20616.
  48. Tang, F., Barbacioru, C., Wang, Y., Nordman, E., Lee, C., Xu, N., Wang, X., Bodeau, J., Tuch, B. B., Siddiqui, A., et al. (2009). mrna-seq whole-transcriptome analysis of a single cell. Nature methods, 6(5):377-382.
  49. Van de Sande, B., Lee, J. S., Mutasa-Gottgens, E., Naughton, B., Bacon, W., Manning, J., Wang, Y., Pollard, J., Mendez, M., Hill, J., Kumar, N., Cao, X., Chen, X., Khaladkar, M., Wen, J., Leach, A., and Ferran, E. (2023). Applications of single-cell RNA sequencing in drug discovery and development. Nat. Rev. Drug Discovery, 22(6):496-520.
  50. Wenteler, A., Occhetta, M., Branson, N., Huebner, M., Curean, V., Dee, W. T., Connell, W. T., Hawkins-Hooker, A., Chung, S. P., Ektefaie, Y., Gallagher-Syed, A., and Córdova, C. M. V. (2024). PertEval-scFM: Benchmarking Single-Cell Foundation Models for Perturbation Effect Prediction. bioRxiv, page 2024.10.02.616248.
  51. Wong, D. R., Hill, A. S., and Moccia, R. (2025). Simple controls exceed best deep learning algorithms and reveal foundation model effectiveness for predicting genetic perturbations. bioRxiv.
  52. Yu, H. and Welch, J. D. (2022). PerturbNet predicts single-cell responses to unseen chemical and genetic perturbations. bioRxiv, page 2022.07.20.500854.
  53. Zhang, J., Ubas, A. A., de Borja, R., Svensson, V., Thomas, N., Thakar, N., Winters, A., Khan, U., Jones, M. G., Thompson, J. D., Tran, V., Pangallo, J., Papalexi, E., Sapre, A., Nguyen, H., Sanderson, O., Nigos, M., Kaplan, O., Schroeder, S., Hariadi, B., Marrujo, S., Salvino, C. C. A., Gallareta Olivares, G., Koehler, R., Geiss, G., Rosenberg, A., Roco, C., Merico, D., Alidoust, N., Goodarzi, H., and Yu, J. (2025). Tahoe-100m: A giga-scale single-cell perturbation atlas for context-dependent gene function and cellular modeling. bioRxiv.

Cited by

Sorted by citations

Authors
Wu, Y., Wershof, E., Schmon, S., Nassar, M., Osiński, B., Eksi, R., Yan, Z., Stark, R., Zhang, K., & Graepel, T.
Affiliations
  • Altos Labs (US)
  • Memorial Sloan Kettering Cancer Center (US)
  • Mohamed bin Zayed University of Artificial Intelligence (AE)
ORCID
Venue
Advances in Neural Information Processing Systems 38
Publisher
Neural Information Processing Systems Foundation, Inc. (NeurIPS)
Published
2025
Pages
106937-106977
ISSN
1049-5258 (ISSN-L)
DOI
10.52202/085713-3225
OpenAlex
W4403007117
Cited by
20
References
0
Citations by year
2026: 13 · 2025: 7
FWCI
78.92
Type
conference paper
Language
en
Open access
Licence
other-oa
Locations
  • neural information processing systems (publishedVersion)
  • arXiv (Cornell University) (submittedVersion, other-oa)
  • arXiv (Cornell University)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Single-cell and spatial transcriptomics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene Regulatory Network Analysis
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene expression and cancer classification
Keywords
single-cell transcriptomics, benchmarking, model evaluation metrics, rank metric, mode collapse, model architecture comparison, genetic screen, chemical screen, therapeutic discovery

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