{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T21:20:31Z","timestamp":1792790431461,"version":"4.5.0"},"posted":{"date-parts":[[2026,9,26]]},"group-title":"Genomics","reference-count":58,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2026,9,26]],"date-time":"2026-09-26T00:00:00Z","timestamp":1790380800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"funder":[{"name":"National Human Genome Research Institute","award":["R21HG012639"],"award-info":[{"award-number":["R21HG012639"]}],"id":[{"id":"https:\/\/ror.org\/00baak391","id-type":"ROR","asserted-by":"crossref"}]},{"award":["2022-253560"],"award-info":[{"award-number":["2022-253560"]}],"id":[{"id":"https:\/\/ror.org\/02qenvm24","id-type":"ROR","asserted-by":"publisher"}]},{"name":"National Institute of Health","award":["R35HG011941"],"award-info":[{"award-number":["R35HG011941"]}]},{"name":"National Institute of Health","award":["R01CA304906"],"award-info":[{"award-number":["R01CA304906"]}]},{"name":"National Institute of Health","award":["GG016982"],"award-info":[{"award-number":["GG016982"]}]},{"name":"National Science Foundation","award":["2146007"],"award-info":[{"award-number":["2146007"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,9,26]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>Understanding the effects of genetic, molecular, and experimental perturbations is essential for decoding cellular mechanisms and guiding biomedical interventions. Existing computational approaches are typically designed for individual tasks, such as predicting perturbation responses, characterizing downstream transcriptional effects, or modeling perturbation combinations, and therefore do not provide a unified framework for learning interpretable perturbation representations while enabling causal analysis of downstream effects and characterization of condition-dependent responses.<\/jats:p>\n                <jats:p>We present SHERLOCK, an interpretable deep generative framework for single-cell perturbation analysis that represents perturbation effects as structured interventions on a latent baseline cellular state. SHERLOCK learns correlated and sparse perturbation representations that organize genetic and pharmacological perturbations according to shared transcriptional responses. By formulating perturbations as interventions within a structural causal model, it enables counterfactual estimation of their downstream transcriptional effects under explicit identifiability assumptions. The same framework quantifies how perturbation responses vary across conditions and compositionally models combinatorial perturbations, enabling prediction of held-out combinations and classification of genetic interactions.<\/jats:p>\n                <jats:p>Across genome-scale CRISPR, chemical, and spatial perturbation datasets, SHERLOCK recovers perturbation relationships concordant with known biological pathways and pharmacological properties, identifies condition-dependent responses, and predicts combinatorial perturbation effects. Together, SHERLOCK provides a unified framework for interpretable and causal analysis of perturbation effects across diverse single-cell perturbation experiments.<\/jats:p>","DOI":"10.64898\/2026.09.25.754573","type":"posted-content","created":{"date-parts":[[2026,9,26]],"date-time":"2026-09-26T16:55:11Z","timestamp":1790441711000},"source":"Crossref","is-referenced-by-count":0,"title":["SHERLOCK: Structured representation learning and causal inference of downstream perturbation effects"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0009-0005-5013-5549","authenticated-orcid":false,"given":"Mingxuan","family":"Zhang","sequence":"first","affiliation":[{"name":"Irving Institute for Cancer Dynamics, Columbia University, New York, NY, USA"},{"name":"Department of Systems Biology, Columbia University Medical Center, New York, NY, USA"},{"name":"Department of Molecular Pharmacology and Therapeutics, Columbia University Medical Center, New York, NY 10032, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Joshua D.","family":"Myers","sequence":"additional","affiliation":[{"name":"Irving Institute for Cancer Dynamics, Columbia University, New York, NY, USA"},{"name":"Department of Biomedical Engineering, Columbia University, New York, NY, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lingting","family":"Shi","sequence":"additional","affiliation":[{"name":"Irving Institute for Cancer Dynamics, Columbia University, New York, NY, USA"},{"name":"Department of Biomedical Engineering, Columbia University, New York, NY, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ross M.","family":"Giglio","sequence":"additional","affiliation":[{"name":"Department of Molecular Pharmacology and Therapeutics, Columbia University Medical Center, New York, NY 10032, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sharanya","family":"Chatterjee","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Columbia University, New York, NY, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jos\u00e9 L.","family":"McFaline-Figueroa","sequence":"additional","affiliation":[{"name":"Irving Institute for Cancer Dynamics, Columbia University, New York, NY, USA"},{"name":"Department of Biomedical Engineering, Columbia University, New York, NY, USA"},{"name":"Herbert Irving Comprehensive Cancer Center, Columbia University, New York, NY, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5059-6971","authenticated-orcid":false,"given":"Elham","family":"Azizi","sequence":"additional","affiliation":[{"name":"Irving Institute for Cancer Dynamics, Columbia University, New York, NY, USA"},{"name":"Department of Biomedical Engineering, Columbia University, New York, NY, USA"},{"name":"Department of Computer Science, Columbia University, New York, NY, USA"},{"name":"Herbert Irving Comprehensive Cancer Center, Columbia University, New York, NY, USA"},{"name":"Data Science Institute, Columbia University, New York, NY, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026092918001570000_2026.09.25.754573v1.1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2016.11.048"},{"key":"2026092918001570000_2026.09.25.754573v1.2","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2016.11.038"},{"key":"2026092918001570000_2026.09.25.754573v1.3","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2016.11.039"},{"key":"2026092918001570000_2026.09.25.754573v1.4","doi-asserted-by":"publisher","DOI":"10.1038\/nmeth.4177"},{"key":"2026092918001570000_2026.09.25.754573v1.5","doi-asserted-by":"publisher","DOI":"10.1126\/science.aax6234"},{"key":"2026092918001570000_2026.09.25.754573v1.6","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2019.09.016"},{"key":"2026092918001570000_2026.09.25.754573v1.7","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2022.02.015"},{"key":"2026092918001570000_2026.09.25.754573v1.8","doi-asserted-by":"publisher","DOI":"10.1038\/s41587-026-03127-y"},{"key":"2026092918001570000_2026.09.25.754573v1.9","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-023-01969-x"},{"key":"2026092918001570000_2026.09.25.754573v1.10","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-019-0494-8"},{"key":"2026092918001570000_2026.09.25.754573v1.11","doi-asserted-by":"publisher","DOI":"10.1038\/s41587-023-01905-6"},{"key":"2026092918001570000_2026.09.25.754573v1.12","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-024-02201-0"},{"key":"2026092918001570000_2026.09.25.754573v1.13","doi-asserted-by":"publisher","DOI":"10.1101\/2023.09.11.557287"},{"key":"2026092918001570000_2026.09.25.754573v1.14","unstructured":"Elias Bareinboim et al. On Pearl\u2019s Hierarchy and the Foundations of Causal Inference. Technical Report R-60. Causal Artificial Intelligence Lab, Columbia University, 2021. url: https:\/\/causalai.net\/r60.pdf."},{"key":"2026092918001570000_2026.09.25.754573v1.15","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2021\/277"},{"key":"2026092918001570000_2026.09.25.754573v1.16","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-023-01955-3"},{"key":"2026092918001570000_2026.09.25.754573v1.17","doi-asserted-by":"crossref","unstructured":"Michael Bereket and Theofanis Karaletsos . \u201cModelling Cellular Perturbations with the Sparse Additive Mechanism Shift Variational Autoencoder\u201d. In: Advances in Neural Information Processing Systems. 2023.","DOI":"10.52202\/075280-0001"},{"key":"2026092918001570000_2026.09.25.754573v1.18","unstructured":"S\u00e9bastien Lachapelle et al. \u201cDisentanglement via Mechanism Sparsity Regularization: A New Principle for Nonlinear ICA\u201d. In: Proceedings of the 1st Conference on Causal Learning and Reasoning (CLeaR). 2022."},{"key":"2026092918001570000_2026.09.25.754573v1.19","unstructured":"Romain Lopez et al. \u201cLearning Causal Representations of Single Cells via Sparse Mechanism Shift Modeling\u201d. In: Proceedings of Machine Learning Research (2023)."},{"key":"2026092918001570000_2026.09.25.754573v1.20","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2022.05.013"},{"key":"2026092918001570000_2026.09.25.754573v1.21","doi-asserted-by":"publisher","DOI":"10.1038\/nrm3350"},{"key":"2026092918001570000_2026.09.25.754573v1.22","doi-asserted-by":"publisher","DOI":"10.1016\/j.bbagrm.2014.03.013"},{"key":"2026092918001570000_2026.09.25.754573v1.23","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2025.09.006"},{"key":"2026092918001570000_2026.09.25.754573v1.24","doi-asserted-by":"publisher","DOI":"10.1101\/2024.04.08.587960"},{"key":"2026092918001570000_2026.09.25.754573v1.25","doi-asserted-by":"publisher","DOI":"10.64898\/2026.01.08.698516"},{"key":"2026092918001570000_2026.09.25.754573v1.26","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-019-0969-x"},{"key":"2026092918001570000_2026.09.25.754573v1.27","doi-asserted-by":"publisher","DOI":"10.1126\/science.aam8999"},{"key":"2026092918001570000_2026.09.25.754573v1.28","doi-asserted-by":"publisher","DOI":"10.1158\/2326-6066.CIR-18-0903"},{"key":"2026092918001570000_2026.09.25.754573v1.29","doi-asserted-by":"publisher","DOI":"10.3389\/fimmu.2019.00860"},{"key":"2026092918001570000_2026.09.25.754573v1.30","doi-asserted-by":"publisher","DOI":"10.1158\/2159-8290.CD-23-0584"},{"key":"2026092918001570000_2026.09.25.754573v1.31","unstructured":"Frank P. Vendetti et al. \u201cATR kinase inhibition promotes antitumor immunity and radiosensitizes tumors\u201d. In: Clinical Cancer Research (2018)."},{"key":"2026092918001570000_2026.09.25.754573v1.32","doi-asserted-by":"publisher","DOI":"10.3389\/fimmu.2021.718621"},{"key":"2026092918001570000_2026.09.25.754573v1.33","doi-asserted-by":"publisher","DOI":"10.1042\/BJ20061844"},{"key":"2026092918001570000_2026.09.25.754573v1.34","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-018-03340-7"},{"key":"2026092918001570000_2026.09.25.754573v1.35","doi-asserted-by":"publisher","DOI":"10.1016\/j.ajpath.2020.12.001"},{"key":"2026092918001570000_2026.09.25.754573v1.36","doi-asserted-by":"publisher","DOI":"10.1101\/gad.344218.120"},{"key":"2026092918001570000_2026.09.25.754573v1.37","doi-asserted-by":"publisher","DOI":"10.1186\/s13046-025-03361-3"},{"key":"2026092918001570000_2026.09.25.754573v1.38","doi-asserted-by":"publisher","DOI":"10.1016\/j.cmet.2023.01.010"},{"key":"2026092918001570000_2026.09.25.754573v1.39","doi-asserted-by":"publisher","DOI":"10.1158\/1541-7786.MCR-16-0234"},{"key":"2026092918001570000_2026.09.25.754573v1.40","doi-asserted-by":"publisher","DOI":"10.1007\/s10238-024-01328-w"},{"key":"2026092918001570000_2026.09.25.754573v1.41","first-page":"1","article-title":"Guanylate-binding proteins: Emerging players in inflammation and cancer","volume":"10","year":"2019","journal-title":"In: Frontiers in Immunology"},{"key":"2026092918001570000_2026.09.25.754573v1.42","doi-asserted-by":"publisher","DOI":"10.3389\/fimmu.2019.03139"},{"key":"2026092918001570000_2026.09.25.754573v1.43","doi-asserted-by":"publisher","DOI":"10.1126\/science.aax4438"},{"key":"2026092918001570000_2026.09.25.754573v1.44","doi-asserted-by":"publisher","DOI":"10.1182\/blood.V45.3.321.321"},{"key":"2026092918001570000_2026.09.25.754573v1.45","doi-asserted-by":"publisher","DOI":"10.1002\/ijc.2910230202"},{"key":"2026092918001570000_2026.09.25.754573v1.46","doi-asserted-by":"publisher","DOI":"10.1101\/gad.13.12.1501"},{"key":"2026092918001570000_2026.09.25.754573v1.47","doi-asserted-by":"publisher","DOI":"10.1038\/sj.onc.1210764"},{"key":"2026092918001570000_2026.09.25.754573v1.48","doi-asserted-by":"publisher","DOI":"10.1016\/j.bbamcr.2007.03.010"},{"key":"2026092918001570000_2026.09.25.754573v1.49","doi-asserted-by":"publisher","DOI":"10.1038\/nri2024"},{"key":"2026092918001570000_2026.09.25.754573v1.50","doi-asserted-by":"publisher","DOI":"10.1101\/gad.13.11.1398"},{"key":"2026092918001570000_2026.09.25.754573v1.51","doi-asserted-by":"publisher","DOI":"10.1128\/MCB.17.4.1947"},{"key":"2026092918001570000_2026.09.25.754573v1.52","doi-asserted-by":"publisher","DOI":"10.1101\/2025.09.08.674897"},{"key":"2026092918001570000_2026.09.25.754573v1.53","doi-asserted-by":"publisher","DOI":"10.1093\/neuonc\/noae085"},{"key":"2026092918001570000_2026.09.25.754573v1.54","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2008.01.025"},{"key":"2026092918001570000_2026.09.25.754573v1.55","doi-asserted-by":"publisher","DOI":"10.1182\/blood-2011-04-285981"},{"key":"2026092918001570000_2026.09.25.754573v1.56","doi-asserted-by":"publisher","DOI":"10.1182\/blood-2011-03-331371"},{"key":"2026092918001570000_2026.09.25.754573v1.57","doi-asserted-by":"publisher","DOI":"10.1111\/j.2044-8295.1910.tb00207.x"},{"key":"2026092918001570000_2026.09.25.754573v1.58","doi-asserted-by":"publisher","DOI":"10.1145\/3292500.3330701"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2026.09.25.754573","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T01:00:26Z","timestamp":1790730026000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.64898\/2026.09.25.754573"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,26]]},"references-count":58,"URL":"https:\/\/doi.org\/10.64898\/2026.09.25.754573","relation":{},"subject":[],"published":{"date-parts":[[2026,9,26]]},"subtype":"preprint"}}