Conference paper · 2025 · Open access

PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis

Yan Wu, Esther Wershof, Sebastian M. Schmon, Marcel Nassar, Błażej Osiński, Ridvan Eksi, Zichao Yan, Rory Stark, Kun Zhang, Thore Graepel

neural information processing systems, 106937–106977 · Published 1 January 2025 · DOI 10.52202/085713-3225

Open PDFView at the publisherCite this paper

In brief

“PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis” is a 2025 conference paper by Yan Wu and 9 co-authors, published in neural information processing systems. OpenAlex counts 102 citations as of 22 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).

Citations
102
Most cited in
2026 (81)
Field-weighted impact
143.56
References
42
Authors
10

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.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Wu, Y., Wershof, E., Schmon, S. M., Nassar, M., Osiński, B., Eksi, R., Yan, Z., Stark, R., Zhang, K., & Graepel, T. (2025). PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis. neural information processing systems, 106937–106977. https://doi.org/10.52202/085713-3225

MLA 9

Wu, Yan, et al. “PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis.” neural information processing systems, 2025, pp. 106937–106977. https://doi.org/10.52202/085713-3225.

Chicago

Wu, Yan, Esther Wershof, Sebastian M. Schmon, Marcel Nassar, Błażej Osiński, Ridvan Eksi, Zichao Yan, Rory Stark, Kun Zhang, and Thore Graepel. 2025. “PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis.” neural information processing systems: 106937–106977. https://doi.org/10.52202/085713-3225.

Harvard

Wu, Y. et al. (2025) ‘PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis’, neural information processing systems, pp. 106937–106977. doi:10.52202/085713-3225.

IEEE

Y. Wu et al., “PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis,” neural information processing systems, pp. 106937–106977, 2025, doi: 10.52202/085713-3225.

BibTeX

@inproceedings{wu2025perturbench,
  title     = {{PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis}},
  author    = {Wu, Yan and Wershof, Esther and Schmon, Sebastian M. and Nassar, Marcel and Osiński, Błażej and Eksi, Ridvan and Yan, Zichao and Stark, Rory and Zhang, Kun and Graepel, Thore},
  booktitle = {neural information processing systems},
  year      = {2025},
  pages     = {106937--106977},
  doi       = {10.52202/085713-3225},
  url       = {https://doi.org/10.52202/085713-3225}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Citations by year

YearCitationsShare of the best year
202411
20252020
20268181

Its field-weighted citation impact is 143.56: it has been cited 143.56 times as often as the average work of the same field, type and year (1.00 is the world average).

Topic and keywords

Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Single-cell and spatial transcriptomics

  • single-cell transcriptomics
  • benchmarking
  • model evaluation metrics
  • rank metric
  • mode collapse
  • model architecture comparison
  • genetic screen
  • chemical screen

Work with this paper

Most-cited papers in Biochemistry, Genetics and Molecular Biology

All highly cited Biochemistry, Genetics and Molecular Biology papers

Questions about this paper

How do I cite this paper in APA style?

Wu, Y., Wershof, E., Schmon, S. M., Nassar, M., Osiński, B., Eksi, R., Yan, Z., Stark, R., Zhang, K., & Graepel, T. (2025). PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis. neural information processing systems, 106937–106977. https://doi.org/10.52202/085713-3225

How many times has this paper been cited?

102 times, according to OpenAlex on 22 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at arXiv (Cornell University): sign in and open the PDF from this page.

Who wrote this paper?

Yan Wu, Esther Wershof, Sebastian M. Schmon, Marcel Nassar, Błażej Osiński, Ridvan Eksi, Zichao Yan, Rory Stark, Kun Zhang, Thore Graepel.

What is the DOI of this paper?

10.52202/085713-3225, permanently at https://doi.org/10.52202/085713-3225. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 22 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.