{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,8,9]],"date-time":"2026-08-09T23:31:46Z","timestamp":1786318306141,"version":"3.56.0"},"posted":{"date-parts":[[2025,6,27]]},"group-title":"Systems Biology","reference-count":120,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2025,6,27]],"date-time":"2025-06-27T00:00:00Z","timestamp":1750982400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2025,7,10]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>\n                  Cellular responses to perturbations are a cornerstone for understanding biological mechanisms and selecting drug targets. While machine learning models offer tremendous potential for predicting perturbation effects, they currently struggle to generalize to unobserved cellular contexts. Here, we introduce S\n                  <jats:sc>tate<\/jats:sc>\n                  , a transformer model that predicts perturbation effects while accounting for cellular heterogeneity within and across experiments. S\n                  <jats:sc>tate<\/jats:sc>\n                  predicts perturbation effects across sets of cells and is trained using gene expression data from over 100 million perturbed cells. S\n                  <jats:sc>tate<\/jats:sc>\n                  improved discrimination of effects on large datasets by more than 30% and identified differentially expressed genes across genetic, signaling and chemical perturbations with significantly improved accuracy. Using its cell embedding trained on observational data from 167 million cells, S\n                  <jats:sc>tate<\/jats:sc>\n                  identified strong perturbations in novel cellular contexts where no perturbations were observed during training. We further introduce Cell-Eval, a comprehensive evaluation framework that highlights S\n                  <jats:sc>tate<\/jats:sc>\n                  \u2019s ability to detect cell type-specific perturbation responses, such as cell survival. Overall, the performance and flexibility of S\n                  <jats:sc>tate<\/jats:sc>\n                  sets the stage for scaling the development of virtual cell models.\n                <\/jats:p>","DOI":"10.1101\/2025.06.26.661135","type":"posted-content","created":{"date-parts":[[2025,6,28]],"date-time":"2025-06-28T03:25:12Z","timestamp":1751081112000},"source":"Crossref","is-referenced-by-count":83,"title":["Predicting cellular responses to perturbation across diverse contexts with State"],"prefix":"10.64898","author":[{"given":"Abhinav K.","family":"Adduri","sequence":"first","affiliation":[{"name":"Arc Institute"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dhruv","family":"Gautam","sequence":"additional","affiliation":[{"name":"Arc Institute"},{"name":"University of California, Berkeley"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Beatrice","family":"Bevilacqua","sequence":"additional","affiliation":[{"name":"Arc Institute"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alishba","family":"Imran","sequence":"additional","affiliation":[{"name":"Arc Institute"},{"name":"University of California, Berkeley"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rohan","family":"Shah","sequence":"additional","affiliation":[{"name":"Arc Institute"},{"name":"University of Pennsylvania"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mohsen","family":"Naghipourfar","sequence":"additional","affiliation":[{"name":"Arc Institute"},{"name":"University of California, Berkeley"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Noam","family":"Teyssier","sequence":"additional","affiliation":[{"name":"Arc Institute"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rajesh","family":"Ilango","sequence":"additional","affiliation":[{"name":"Arc Institute"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sanjay","family":"Nagaraj","sequence":"additional","affiliation":[{"name":"Arc Institute"},{"name":"Stanford University;"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mingze","family":"Dong","sequence":"additional","affiliation":[{"name":"Arc Institute"},{"name":"Yale University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chiara","family":"Ricci-Tam","sequence":"additional","affiliation":[{"name":"Arc Institute"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christopher","family":"Carpenter","sequence":"additional","affiliation":[{"name":"Arc Institute"},{"name":"University of California, San Francisco"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Vishvak","family":"Subramanyam","sequence":"additional","affiliation":[{"name":"Arc Institute"},{"name":"University of California, San Francisco"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aidan","family":"Winters","sequence":"additional","affiliation":[{"name":"Arc Institute"},{"name":"University of California, San Francisco"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sravya","family":"Tirukkovular","sequence":"additional","affiliation":[{"name":"Arc Institute"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jeremy","family":"Sullivan","sequence":"additional","affiliation":[{"name":"Arc Institute"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Brian S.","family":"Plosky","sequence":"additional","affiliation":[{"name":"Arc Institute"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Basak","family":"Eraslan","sequence":"additional","affiliation":[{"name":"Arc Institute"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nicholas D.","family":"Youngblut","sequence":"additional","affiliation":[{"name":"Arc Institute"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jure","family":"Leskovec","sequence":"additional","affiliation":[{"name":"Stanford University;"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Luke A.","family":"Gilbert","sequence":"additional","affiliation":[{"name":"Arc Institute"},{"name":"University of California, San Francisco"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Silvana","family":"Konermann","sequence":"additional","affiliation":[{"name":"Arc Institute"},{"name":"Stanford University;"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Patrick D.","family":"Hsu","sequence":"additional","affiliation":[{"name":"Arc Institute"},{"name":"University of California, Berkeley"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alexander","family":"Dobin","sequence":"additional","affiliation":[{"name":"Arc Institute"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dave P.","family":"Burke","sequence":"additional","affiliation":[{"name":"Arc Institute"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9648-8949","authenticated-orcid":false,"given":"Hani","family":"Goodarzi","sequence":"additional","affiliation":[{"name":"Arc Institute"},{"name":"University of California, San Francisco"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9577-5014","authenticated-orcid":false,"given":"Yusuf H.","family":"Roohani","sequence":"additional","affiliation":[{"name":"Arc Institute"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026041210453674000_2025.06.26.661135v2.1","doi-asserted-by":"crossref","unstructured":"C. Ahlmann-Eltze , W. Huber , and S. Anders . Deep learning-based predictions of gene perturbation effects do not yet outperform simple linear methods. BioRxiv, pages 2024\u201309, 2024.","DOI":"10.1101\/2024.09.16.613342"},{"key":"2026041210453674000_2025.06.26.661135v2.2","unstructured":"E. Ameisen , J. Lindsey , A. Pearce , W. Gurnee , N. L. Turner , B. Chen , C. Citro , D. Abrahams , S. Carter , B. Hosmer , J. Marcus , M. Sklar , A. Templeton , T. Bricken , C. McDougall , H. Cunningham , T. Henighan , A. Jermyn , A. Jones , A. Persic , Z. Qi , T. Ben Thompson , S. Zimmerman , K. Rivoire , T. Conerly , C. Olah , and J. Batson . Circuit tracing: Revealing computational graphs in language models. Transformer Circuits Thread, 2025."},{"key":"2026041210453674000_2025.06.26.661135v2.3","doi-asserted-by":"publisher","DOI":"10.1016\/j.tiv.2021.105108"},{"key":"2026041210453674000_2025.06.26.661135v2.4","doi-asserted-by":"crossref","unstructured":"D. Bau , B. Zhou , A. Khosla , A. Oliva , and A. Torralba . Network dissection: Quantifying interpretability of deep visual representations. In Proceedings of the IEEE conference on computer vision and pattern recognition, 2017.","DOI":"10.1109\/CVPR.2017.354"},{"key":"2026041210453674000_2025.06.26.661135v2.5","doi-asserted-by":"crossref","unstructured":"M. Bereket and T. Karaletsos . Modelling cellular perturbations with the sparse additive mechanism shift variational autoencoder. Advances in Neural Information Processing Systems,36, 2024.","DOI":"10.52202\/075280-0001"},{"key":"2026041210453674000_2025.06.26.661135v2.6","doi-asserted-by":"crossref","unstructured":"A. Bhaskar , A. Wettig , D. Friedman , and D. Chen . Finding transformer circuits with edge pruning. Advances in Neural Information Processing Systems, 37, 2024.","DOI":"10.52202\/079017-0587"},{"key":"2026041210453674000_2025.06.26.661135v2.7","article-title":"Identification of mean-field dynamics using transformers","year":"2024","journal-title":"arXiv preprint"},{"key":"2026041210453674000_2025.06.26.661135v2.8","doi-asserted-by":"publisher","DOI":"10.1038\/s43586-022-00144-4"},{"key":"2026041210453674000_2025.06.26.661135v2.9","doi-asserted-by":"crossref","unstructured":"P. Boyeau , J. Hong , A. Gayoso , M. Kim , J. L. McFaline-Figueroa , M. I. Jordan , E. Azizi , C. Ergen , and N. Yosef . Deep generative modeling of sample-level heterogeneity in singlecell genomics. BioRxiv, pages 2022\u201310, 2022.","DOI":"10.1101\/2022.10.04.510898"},{"key":"2026041210453674000_2025.06.26.661135v2.10","doi-asserted-by":"publisher","DOI":"10.1002\/cpa.3160440402"},{"key":"2026041210453674000_2025.06.26.661135v2.11","unstructured":"T. Bricken , A. Templeton , J. Batson , B. Chen , A. Jermyn , T. Conerly , N. Turner , C. Anil , C. Denison , A. Askell , R. Lasenby , Y. Wu , S. Kravec , N. Schiefer , T. Maxwell , N. Joseph , Z. Hatfield-Dodds , A. Tamkin , K. Nguyen , B. McLean , J. E. Burke , T. Hume , S. Carter , T. Henighan , and C. Olah . Towards monosemanticity: Decomposing language models with dictionary learning. Transformer Circuits Thread, 2023."},{"key":"2026041210453674000_2025.06.26.661135v2.12","doi-asserted-by":"crossref","unstructured":"G. Brixi , M. G. Durrant , J. Ku , M. Poli , G. Brockman , D. Chang , G. A. Gonzalez , S. H. King , D. B. Li , A. T. Merchant , et al. Genome modeling and design across all domains of life with evo 2. BioRxiv, pages 2025\u201302, 2025.","DOI":"10.1101\/2025.02.18.638918"},{"key":"2026041210453674000_2025.06.26.661135v2.13","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-023-01969-x"},{"key":"2026041210453674000_2025.06.26.661135v2.14","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2024.11.015"},{"key":"2026041210453674000_2025.06.26.661135v2.15","doi-asserted-by":"publisher","DOI":"10.1038\/s43586-024-00334-2"},{"key":"2026041210453674000_2025.06.26.661135v2.16","doi-asserted-by":"publisher","DOI":"10.2307\/2152752"},{"key":"2026041210453674000_2025.06.26.661135v2.17","doi-asserted-by":"crossref","unstructured":"H. Chefer , S. Gur , and L. Wolf . Transformer interpretability beyond attention visualization. In Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition, pages 782\u2013791, 2021.","DOI":"10.1109\/CVPR46437.2021.00084"},{"key":"2026041210453674000_2025.06.26.661135v2.18","doi-asserted-by":"crossref","unstructured":"Y. Chen and J. Zou . Genept: a simple but effective foundation model for genes and cells built from chatgpt. bioRxiv, pages 2023\u201310, 2024.","DOI":"10.1101\/2023.10.16.562533"},{"key":"2026041210453674000_2025.06.26.661135v2.19","article-title":"Fast and scalable wasserstein-1 neural optimal transport solver for single-cell perturbation prediction","year":"2024","journal-title":"arXiv preprint"},{"key":"2026041210453674000_2025.06.26.661135v2.20","article-title":"Causalbench: A largescale benchmark for network inference from single-cell perturbation data","year":"2022","journal-title":"arXiv preprint"},{"key":"2026041210453674000_2025.06.26.661135v2.21","article-title":"What does bert look at? an analysis of bert\u2019s attention","year":"2019","journal-title":"arXiv preprint"},{"key":"2026041210453674000_2025.06.26.661135v2.22","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-024-02201-0"},{"key":"2026041210453674000_2025.06.26.661135v2.23","unstructured":"M. Cuturi . Sinkhorn distances: Lightspeed computation of optimal transport. Advances in neural information processing systems, 26, 2013."},{"key":"2026041210453674000_2025.06.26.661135v2.24","article-title":"Provable optimal transport with transformers: The essence of depth and prompt engineering","year":"2024","journal-title":"arXiv preprint"},{"key":"2026041210453674000_2025.06.26.661135v2.25","doi-asserted-by":"publisher","DOI":"10.1038\/s41587-021-01033-z"},{"key":"2026041210453674000_2025.06.26.661135v2.26","unstructured":"T. Dao . Flashattention-2: Faster attention with better parallelism and work partitioning. In The Twelfth International Conference on Learning Representations, 2024."},{"key":"2026041210453674000_2025.06.26.661135v2.27","doi-asserted-by":"publisher","DOI":"10.1038\/nmeth.4177"},{"key":"2026041210453674000_2025.06.26.661135v2.28","unstructured":"V. De la Pena and E. Gin\u00e9 . Decoupling: from dependence to independence. Springer Science & Business Media, 2012."},{"key":"2026041210453674000_2025.06.26.661135v2.29","article-title":"sc-otgm: Single-cell perturbation modeling by solving optimal mass transport on the manifold of gaussian mixtures","year":"2024","journal-title":"arXiv preprint"},{"key":"2026041210453674000_2025.06.26.661135v2.30","doi-asserted-by":"crossref","unstructured":"J. Ding , J. Lin , S. Jiang , Y. Wang , Z. Miao , Z. Fang , J. Tang , M. Li , and X. Qiu . Toward a privacy-preserving predictive foundation model of single-cell transcriptomics with federated learning and tabular modeling. bioRxiv, pages 2025\u201301, 2025.","DOI":"10.1101\/2025.01.06.631427"},{"key":"2026041210453674000_2025.06.26.661135v2.31","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2016.11.038"},{"key":"2026041210453674000_2025.06.26.661135v2.32","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-023-02040-5"},{"key":"2026041210453674000_2025.06.26.661135v2.33","doi-asserted-by":"crossref","unstructured":"J. Dunefsky , P. Chlenski , and N. Nanda . Transcoders find interpretable llm feature circuits. In The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024.","DOI":"10.52202\/079017-0768"},{"key":"2026041210453674000_2025.06.26.661135v2.34","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2017.12.012"},{"key":"2026041210453674000_2025.06.26.661135v2.35","doi-asserted-by":"crossref","unstructured":"C. Feng , E. M. Peets , Y. Zhou , L. Crepaldi , S. Usluer , A. Dunham , J. M. Braunger , J. Su , M. E. Strauss , D. Muraro , et al. A genome-scale single cell crispri map of trans gene regulation across human pluripotent stem cell lines. bioRxiv, pages 2024\u201311, 2024.","DOI":"10.1101\/2024.11.28.625833"},{"key":"2026041210453674000_2025.06.26.661135v2.36","unstructured":"J. Feydy , T. S\u00e9journ\u00e9 , F.-X. Vialard , S.-i. Amari , A. Trouve , and G. Peyr\u00e9 . Interpolating between optimal transport and mmd using sinkhorn divergences. In The 22nd International Conference on Artificial Intelligence and Statistics, pages 2681\u20132690, 2019."},{"issue":"1","key":"2026041210453674000_2025.06.26.661135v2.37","doi-asserted-by":"crossref","first-page":"6611","DOI":"10.1038\/s41467-024-51059-5","article-title":"sctab: scaling cross-tissue single-cell annotation models","volume":"15","year":"2024","journal-title":"Nature Communications"},{"key":"2026041210453674000_2025.06.26.661135v2.38","article-title":"Transformers are universal in-context learners","year":"2024","journal-title":"arXiv preprint"},{"issue":"25","key":"2026041210453674000_2025.06.26.661135v2.39","first-page":"723","article-title":"A kernel two-sample test","volume":"13","year":"2012","journal-title":"Journal of Machine Learning Research"},{"key":"2026041210453674000_2025.06.26.661135v2.40","unstructured":"S. Gunasekar , J. Lee , D. Soudry , and N. Srebro . Characterizing implicit bias in terms of optimization geometry. In International Conference on Machine Learning, pages 1832\u2013 1841. PMLR, 2018."},{"key":"2026041210453674000_2025.06.26.661135v2.41","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-024-02305-7"},{"key":"2026041210453674000_2025.06.26.661135v2.42","doi-asserted-by":"crossref","unstructured":"P. Hase , M. Bansal , B. Kim , and A. Ghandeharioun . Does localization inform editing? surprising differences in causality-based localization vs. knowledge editing in language models. Advances in Neural Information Processing Systems, 36, 2023.","DOI":"10.52202\/075280-0774"},{"key":"2026041210453674000_2025.06.26.661135v2.43","doi-asserted-by":"crossref","unstructured":"T. Hayes , R. Rao , H. Akin , N. J. Sofroniew , D. Oktay , Z. Lin , R. Verkuil , V. Q. Tran , J. Deaton , M. Wiggert , et al. Simulating 500 million years of evolution with a language model. Science, page eads0018, 2025.","DOI":"10.1101\/2024.07.01.600583"},{"key":"2026041210453674000_2025.06.26.661135v2.44","doi-asserted-by":"crossref","unstructured":"K. He , X. Zhang , S. Ren , and J. Sun . Delving deep into rectifiers: Surpassing humanlevel performance on imagenet classification. In Proceedings of the IEEE international conference on computer vision, pages 1026\u20131034, 2015.","DOI":"10.1109\/ICCV.2015.123"},{"issue":"4","key":"2026041210453674000_2025.06.26.661135v2.45","doi-asserted-by":"crossref","first-page":"239","DOI":"10.1016\/j.cels.2016.04.001","article-title":"Low dimensionality in gene expression data enables the accurate extraction of transcriptional programs from shallow sequencing","volume":"2","year":"2016","journal-title":"Cell systems"},{"key":"2026041210453674000_2025.06.26.661135v2.46","doi-asserted-by":"crossref","unstructured":"N. Ho , C. N. Ellington , J. Hou , S. Addagudi , S. Mo , T. Tao , D. Li , Y. Zhuang , H. Wang , X. Cheng , et al. Scaling dense representations for single cell with transcriptome-scale context. bioRxiv, pages 2024\u201311, 2024.","DOI":"10.1101\/2024.11.28.625303"},{"key":"2026041210453674000_2025.06.26.661135v2.47","unstructured":"W. Hoeffding . The strong law of large numbers for u-statistics. Univ. of North Carolina Institute of statistics. Mimeo Series, 302, 1961."},{"key":"2026041210453674000_2025.06.26.661135v2.48","doi-asserted-by":"crossref","unstructured":"A. C. Huang , T.-H. S. Hsieh , J. Zhu , J. Michuda , A. Teng , S. Kim , E. M. Rumsey , S. K. Lam , I. Anigbogu , P. Wright , et al. X-atlas\/orion: Genome-wide perturb-seq datasets via a scalable fix-cryopreserve platform for training dose-dependent biological foundation models. bioRxiv, pages 2025\u201306, 2025a.","DOI":"10.1101\/2025.06.11.659105"},{"key":"2026041210453674000_2025.06.26.661135v2.49","doi-asserted-by":"crossref","unstructured":"K. Huang , S. Zhang , H. Wang , Y. Qu , Y. Lu , Y. Roohani , R. Li , L. Qiu , J. Zhang , Y. Di , et al. Biomni: A general-purpose biomedical ai agent. bioRxiv, pages 2025\u201305, 2025b.","DOI":"10.1101\/2025.05.30.656746"},{"issue":"6","key":"2026041210453674000_2025.06.26.661135v2.50","doi-asserted-by":"crossref","first-page":"522","DOI":"10.1016\/j.cels.2021.05.016","article-title":"Machine learning for perturbational single-cell omics","volume":"12","year":"2021","journal-title":"Cell Systems"},{"key":"2026041210453674000_2025.06.26.661135v2.51","doi-asserted-by":"crossref","unstructured":"J. Jiang , C. Wang , R. Qi , H. Fu , and Q. Ma . scread: a single-cell rna-seq database for alzheimer\u2019s disease. Iscience, 23(11), 2020.","DOI":"10.1016\/j.isci.2020.101769"},{"key":"2026041210453674000_2025.06.26.661135v2.52","doi-asserted-by":"crossref","unstructured":"L. Jiang , C. Dalgarno , E. Papalexi , I. Mascio , H.-H. Wessels , H. Yun , N. Iremadze , G. Lithwick-Yanai , D. Lipson , and R. Satija . Systematic reconstruction of molecular pathway signatures using scalable single-cell perturbation screens. Nature Cell Biology, pages 1\u201313, 2025.","DOI":"10.1101\/2024.01.29.576933"},{"issue":"5","key":"2026041210453674000_2025.06.26.661135v2.53","doi-asserted-by":"crossref","first-page":"btae265","DOI":"10.1093\/bioinformatics\/btae265","article-title":"scpram accurately predicts single-cell gene expression perturbation response based on attention mechanism","volume":"40","year":"2024","journal-title":"Bioinformatics"},{"key":"2026041210453674000_2025.06.26.661135v2.54","article-title":"A study of bfloat16 for deep learning training","year":"2019","journal-title":"arXiv preprint"},{"key":"2026041210453674000_2025.06.26.661135v2.55","doi-asserted-by":"crossref","unstructured":"E. Kernfeld , Y. Yang , J. S. Weinstock , A. Battle , and P. Cahan . A systematic comparison of computational methods for expression forecasting. BioRxiv, pages 2023\u201307, 2023.","DOI":"10.1101\/2023.07.28.551039"},{"key":"2026041210453674000_2025.06.26.661135v2.56","article-title":"Zclip: Adaptive spike mitigation for llm pre-training","year":"2025","journal-title":"arXiv preprint"},{"key":"2026041210453674000_2025.06.26.661135v2.57","doi-asserted-by":"crossref","unstructured":"A. Lal , A. Karollus , L. Gunsalus , D. Garfield , S. Nair , A. M. Tseng , M. G. Gordon , J. Blischak , B. van de Geijn , T. Bhangale , et al. Decoding sequence determinants of gene expression in diverse cellular and disease states. bioRxiv, pages 2024\u201310, 2024.","DOI":"10.1101\/2024.10.09.617507"},{"key":"2026041210453674000_2025.06.26.661135v2.58","doi-asserted-by":"publisher","DOI":"10.1038\/s41588-023-01471-2"},{"key":"2026041210453674000_2025.06.26.661135v2.59","unstructured":"J. Lee , Y. Lee , J. Kim , A. Kosiorek , S. Choi , and Y. W. Teh . Set transformer: A framework for attention-based permutation-invariant neural networks. In International conference on machine learning, pages 3744\u20133753. PMLR, 2019."},{"key":"2026041210453674000_2025.06.26.661135v2.60","unstructured":"C. Li , H. Gao , Y. She , H. Bian , Q. Chen , K. Liu , L. Wei , and X. Zhang . Benchmarking ai models for in silico gene perturbation of cells. bioRxiv, pages 2024\u201312, 2024a."},{"key":"2026041210453674000_2025.06.26.661135v2.61","doi-asserted-by":"crossref","unstructured":"L. Li , Y. You , W. Liao , X. Fan , S. Lu , Y. Cao , B. Li , W. Ren , Y. Fu , J. Kong , et al. A systematic comparison of single-cell perturbation response prediction models. bioRxiv, pages 2024\u201312, 2024b.","DOI":"10.1101\/2024.12.23.630036"},{"key":"2026041210453674000_2025.06.26.661135v2.62","doi-asserted-by":"publisher","DOI":"10.1126\/science.ade2574"},{"key":"2026041210453674000_2025.06.26.661135v2.63","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-018-0229-2"},{"key":"2026041210453674000_2025.06.26.661135v2.64","unstructured":"R. Lopez , N. Tagasovska , S. Ra , K. Cho , J. Pritchard , and A. Regev . Learning causal representations of single cells via sparse mechanism shift modeling. In Conference on Causal Learning and Reasoning, pages 662\u2013691. PMLR, 2023."},{"key":"2026041210453674000_2025.06.26.661135v2.65","unstructured":"I. Loshchilov and F. Hutter . Decoupled weight decay regularization. In International Conference on Learning Representations, 2017."},{"key":"2026041210453674000_2025.06.26.661135v2.66","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-019-0494-8"},{"key":"2026041210453674000_2025.06.26.661135v2.67","doi-asserted-by":"publisher","DOI":"10.15252\/msb.202211517"},{"key":"2026041210453674000_2025.06.26.661135v2.68","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-021-01336-8"},{"key":"2026041210453674000_2025.06.26.661135v2.69","unstructured":"M. D. Luecken , S. Gigante , D. B. Burkhardt , R. Cannoodt , D. C. Strobl , N. S. Markov , L. Zappia , G. Palla , W. Lewis , D. Dimitrov , et al. Defining and benchmarking open problems in single-cell analysis. Nature Biotechnology, pages 1\u20136, 2025."},{"key":"2026041210453674000_2025.06.26.661135v2.70","doi-asserted-by":"crossref","unstructured":"I. Lugowska , H. Kose\u0142a-Paterczyk , K. Kozak , and P. Rutkowski . Trametinib: a mek inhibitor for management of metastatic melanoma. OncoTargets and therapy, pages 2251\u20132259, 2015.","DOI":"10.2147\/OTT.S72951"},{"key":"2026041210453674000_2025.06.26.661135v2.71","unstructured":"C. Lyle , Z. Zheng , K. Khetarpal , J. Martens , H. van Hasselt , R. Pascanu , and W. Dabney . Normalization and effective learning rates in reinforcement learning, 2024. URL https:\/\/arxiv.org\/abs\/2407.01800."},{"key":"2026041210453674000_2025.06.26.661135v2.72","unstructured":"A. Makkuva , A. Taghvaei , S. Oh , and J. Lee . Optimal transport mapping via input convex neural networks. In International Conference on Machine Learning, pages 6672\u20136681. PMLR, 2020."},{"key":"2026041210453674000_2025.06.26.661135v2.73","doi-asserted-by":"publisher","DOI":"10.1038\/s41588-019-0489-5"},{"key":"2026041210453674000_2025.06.26.661135v2.74","doi-asserted-by":"crossref","unstructured":"J. L. McFaline-Figueroa , S. Srivatsan , A. J. Hill , M. Gasperini , D. L. Jackson , L. Saunders , S. Domcke , S. G. Regalado , P. Lazarchuck , S. Alvarez , et al. Multiplex single-cell chemical genomics reveals the kinase dependence of the response to targeted therapy. Cell Genomics, 4(2), 2024.","DOI":"10.1016\/j.xgen.2023.100487"},{"key":"2026041210453674000_2025.06.26.661135v2.75","doi-asserted-by":"crossref","unstructured":"A. Nadig , J. M. Replogle , A. N. Pogson , M. Murthy , S. A. McCarroll , J. S. Weissman , E. B. Robinson , and L. J. O\u2019Connor. Transcriptome-wide analysis of differential expression in perturbation atlases. Nature Genetics, pages 1\u201310, 2025.","DOI":"10.1038\/s41588-025-02169-3"},{"key":"2026041210453674000_2025.06.26.661135v2.76","doi-asserted-by":"publisher","DOI":"10.1126\/science.ado9336"},{"key":"2026041210453674000_2025.06.26.661135v2.77","doi-asserted-by":"publisher","DOI":"10.1126\/science.aax4438"},{"key":"2026041210453674000_2025.06.26.661135v2.78","article-title":"Virtual cells: Predict, explain, discover","year":"2025","journal-title":"arXiv preprint"},{"key":"2026041210453674000_2025.06.26.661135v2.79","doi-asserted-by":"publisher","DOI":"10.23915\/distill.00007"},{"key":"2026041210453674000_2025.06.26.661135v2.80","doi-asserted-by":"publisher","DOI":"10.1038\/s41588-021-00778-2"},{"key":"2026041210453674000_2025.06.26.661135v2.81","unstructured":"Parse Biosciences. 10 million human pbmcs in a single experiment, 2023. URL https:\/\/www.parsebiosciences.com\/datasets\/10-million-human-pbmcs-in-a-single-experiment\/."},{"key":"2026041210453674000_2025.06.26.661135v2.82","doi-asserted-by":"crossref","unstructured":"J. D. Pearce , S. E. Simmonds , G. Mahmoudabadi , L. Krishnan , G. Palla , A.-M. Istrate , A. Tarashansky , B. Nelson , O. Valenzuela , D. Li , et al. A cross-species generative cell atlas across 1.5 billion years of evolution: The transcriptformer single-cell model. bioRxiv, pages 2025\u201304, 2025.","DOI":"10.1101\/2025.04.25.650731"},{"key":"2026041210453674000_2025.06.26.661135v2.83","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-023-02144-y"},{"key":"2026041210453674000_2025.06.26.661135v2.84","doi-asserted-by":"publisher","DOI":"10.1038\/s41587-023-01716-9"},{"key":"2026041210453674000_2025.06.26.661135v2.85","doi-asserted-by":"publisher","DOI":"10.1561\/2200000073"},{"key":"2026041210453674000_2025.06.26.661135v2.86","doi-asserted-by":"crossref","unstructured":"Z. Piran , N. Cohen , Y. Hoshen , and M. Nitzan . Disentanglement of single-cell data with biolord. Nature Biotechnology, pages 1\u20136, 2024.","DOI":"10.1038\/s41587-023-02079-x"},{"key":"2026041210453674000_2025.06.26.661135v2.87","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkae1142"},{"key":"2026041210453674000_2025.06.26.661135v2.88","doi-asserted-by":"publisher","DOI":"10.1038\/s41576-021-00409-w"},{"issue":"8","key":"2026041210453674000_2025.06.26.661135v2.89","first-page":"9","article-title":"Language models are unsupervised multitask learners","volume":"1","year":"2019","journal-title":"OpenAI blog"},{"key":"2026041210453674000_2025.06.26.661135v2.90","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2022.05.013"},{"key":"2026041210453674000_2025.06.26.661135v2.91","doi-asserted-by":"crossref","unstructured":"S. A. Rizvi , D. Levine , A. Patel , S. Zhang , E. Wang , S. He , D. Zhang , C. Tang , Z. Lyu , R. Darji , et al. Scaling large language models for next-generation single-cell analysis. bioRxiv, pages 2025\u201304, 2025.","DOI":"10.1101\/2025.04.14.648850"},{"key":"2026041210453674000_2025.06.26.661135v2.92","doi-asserted-by":"publisher","DOI":"10.1002\/wics.1375"},{"key":"2026041210453674000_2025.06.26.661135v2.93","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2024.07.035"},{"key":"2026041210453674000_2025.06.26.661135v2.94","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-024-08338-4"},{"key":"2026041210453674000_2025.06.26.661135v2.95","doi-asserted-by":"publisher","DOI":"10.1038\/s41587-023-01905-6"},{"key":"2026041210453674000_2025.06.26.661135v2.96","article-title":"Biodiscoveryagent: An ai agent for designing genetic perturbation experiments","year":"2024","journal-title":"arXiv preprint"},{"key":"2026041210453674000_2025.06.26.661135v2.97","doi-asserted-by":"crossref","unstructured":"Y. Rosen , Y. Roohani , A. Agarwal , L. Samotor\u010dan , T. S. Consortium , S. R. Quake , and J. Leskovec . Universal cell embeddings: A foundation model for cell biology. bioRxiv, pages 2023\u201311, 2023.","DOI":"10.1101\/2023.11.28.568918"},{"key":"2026041210453674000_2025.06.26.661135v2.98","article-title":"Cross-modality matching and prediction of perturbation responses with labeled gromov-wasserstein optimal transport","year":"2024","journal-title":"arXiv preprint"},{"key":"2026041210453674000_2025.06.26.661135v2.99","unstructured":"R. A. Saunders , W. E. Allen , X. Pan , J. Sandhu , J. Lu , T. K. Lau , K. Smolyar , Z. A. Sullivan , C. Dulac , J. S. Weissman , et al. A platform for multimodal in vivo pooled genetic screens reveals regulators of liver function. bioRxiv, pages 2024\u201311, 2024."},{"key":"2026041210453674000_2025.06.26.661135v2.100","doi-asserted-by":"crossref","unstructured":"B. Song , D. Liu , W. Dai , N. F. McMyn , Q. Wang , D. Yang , A. Krejci , A. Vasilyev , N. Untermoser , A. Loregger , et al. Decoding heterogeneous single-cell perturbation responses. Nature cell biology, pages 1\u201312, 2025.","DOI":"10.1038\/s41556-025-01626-9"},{"issue":"70","key":"2026041210453674000_2025.06.26.661135v2.101","first-page":"1","article-title":"The implicit bias of gradient descent on separable data","volume":"19","year":"2018","journal-title":"Journal of Machine Learning Research"},{"key":"2026041210453674000_2025.06.26.661135v2.102","doi-asserted-by":"publisher","DOI":"10.1126\/science.aax6234"},{"key":"2026041210453674000_2025.06.26.661135v2.103","doi-asserted-by":"crossref","unstructured":"H. Steck , C. Ekanadham , and N. Kallus . Is cosine-similarity of embeddings really about similarity? In Companion Proceedings of the ACM Web Conference 2024, pages 887\u2013890, 2024.","DOI":"10.1145\/3589335.3651526"},{"issue":"8","key":"2026041210453674000_2025.06.26.661135v2.104","doi-asserted-by":"crossref","first-page":"1430","DOI":"10.1038\/s41592-024-02353-z","article-title":"Transformers in single-cell omics: a review and new perspectives","volume":"21","year":"2024","journal-title":"Nature methods"},{"key":"2026041210453674000_2025.06.26.661135v2.105","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06139-9"},{"key":"2026041210453674000_2025.06.26.661135v2.106","article-title":"Llama: Open and efficient foundation language models","year":"2023","journal-title":"arXiv preprint"},{"issue":"6","key":"2026041210453674000_2025.06.26.661135v2.107","doi-asserted-by":"crossref","first-page":"86","DOI":"10.1145\/3571070","article-title":"On the implicit bias in deep-learning algorithms","volume":"66","year":"2023","journal-title":"Communications of the ACM"},{"key":"2026041210453674000_2025.06.26.661135v2.108","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-023-01955-3"},{"key":"2026041210453674000_2025.06.26.661135v2.109","article-title":"Modeling variable guide efficiency in pooled crispr screens with contrastivevi+","year":"2024","journal-title":"arXiv preprint"},{"key":"2026041210453674000_2025.06.26.661135v2.110","doi-asserted-by":"crossref","unstructured":"A. Wenteler , M. Occhetta , N. Branson , M. Huebner , V. Curean , W. Dee , W. Connell , A. Hawkins-Hooker , P. Chung , Y. Ektefaie , et al. Perteval-scfm: Benchmarking single-cell foundation models for perturbation effect prediction. bioRxiv, pages 2024\u201310, 2024.","DOI":"10.1101\/2024.10.02.616248"},{"key":"2026041210453674000_2025.06.26.661135v2.111","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gks1114"},{"key":"2026041210453674000_2025.06.26.661135v2.112","article-title":"Contextualizing biological perturbation experiments through language","year":"2025","journal-title":"arXiv preprint"},{"key":"2026041210453674000_2025.06.26.661135v2.113","article-title":"Perturbench: Benchmarking machine learning models for cellular perturbation analysis","year":"2024","journal-title":"arXiv preprint"},{"key":"2026041210453674000_2025.06.26.661135v2.114","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btu613"},{"key":"2026041210453674000_2025.06.26.661135v2.115","doi-asserted-by":"crossref","unstructured":"N. D. Youngblut , C. Carpenter , J. Prashar , C. Ricci-Tam , R. Ilango , N. Teyssier , S. Konermann , P. Hsu , A. Dobin , D. P. Burke , et al. scBaseCount: An AI agent-curated, uniformly processed, and continually expanding single cell data repository. bioRxiv, pages 2025\u201302, 2025.","DOI":"10.1101\/2025.02.27.640494"},{"key":"2026041210453674000_2025.06.26.661135v2.116","doi-asserted-by":"publisher","DOI":"10.1101\/2024.12.13.628239"},{"key":"2026041210453674000_2025.06.26.661135v2.117","unstructured":"M. Zaheer , S. Kottur , S. Ravanbakhsh , B. Poczos , R. R. Salakhutdinov , and A. J. Smola . Deep sets. Advances in neural information processing systems, 30, 2017."},{"key":"2026041210453674000_2025.06.26.661135v2.118","doi-asserted-by":"crossref","unstructured":"M. D. Zeiler and R. Fergus . Visualizing and understanding convolutional networks. In Computer Vision\u2013ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part I 13, pages 818\u2013833. Springer, 2014.","DOI":"10.1007\/978-3-319-10590-1_53"},{"key":"2026041210453674000_2025.06.26.661135v2.119","first-page":"50254","article-title":"Identifiability guarantees for causal disentanglement from soft interventions","volume":"36","year":"2023","journal-title":"Advances in Neural Information Processing Systems"},{"key":"2026041210453674000_2025.06.26.661135v2.120","doi-asserted-by":"crossref","unstructured":"J. Zhang , A. A. Ubas , R. de Borja , V. Svensson , N. Thomas , N. Thakar , I. Lai , A. Winters , U. Khan , M. G. Jones , J. D. Thompson , V. Tran , J. Pangallo , E. Papalexi , A. Sapre , H. Nguyen , O. Sanderson , M. Nigos , O. Kaplan , S. Schroeder , B. Hariadi , S. Marrujo , C. C. A. Salvino , G. Gallareta Olivares , R. Koehler , G. Geiss , A. Rosenberg , C. Roco , D. Merico , N. Alidoust , H. Goodarzi , and J. Yu . Tahoe-100m: A giga-scale single-cell perturbation atlas for context-dependent gene function and cellular modeling. bioRxiv, pages 2025\u201302, 2025.","DOI":"10.1101\/2025.02.20.639398"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1101\/2025.06.26.661135","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,12]],"date-time":"2026-04-12T17:45:54Z","timestamp":1776015954000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.1101\/2025.06.26.661135"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,27]]},"references-count":120,"URL":"https:\/\/doi.org\/10.1101\/2025.06.26.661135","relation":{},"subject":[],"published":{"date-parts":[[2025,6,27]]},"subtype":"preprint"}}