{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T10:51:51Z","timestamp":1785408711693,"version":"3.56.0"},"posted":{"date-parts":[[2025,9,30]]},"group-title":"Synthetic Biology","reference-count":72,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2025,9,30]],"date-time":"2025-09-30T00:00:00Z","timestamp":1759190400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"name":"NIH","award":["R33CA286947"],"award-info":[{"award-number":["R33CA286947"]}]},{"name":"NIH","award":["R01GM149631"],"award-info":[{"award-number":["R01GM149631"]}]},{"name":"NIH","award":["R56HG013312"],"award-info":[{"award-number":["R56HG013312"]}]},{"name":"NIH","award":["R35HG011317"],"award-info":[{"award-number":["R35HG011317"]}]},{"name":"NIH","award":["R01EY028584"],"award-info":[{"award-number":["R01EY028584"]}]},{"name":"NIH","award":["R01EY033364"],"award-info":[{"award-number":["R01EY033364"]}]},{"award":["Postdoctoral Fellowship in Macular Degeneration"],"award-info":[{"award-number":["Postdoctoral Fellowship in Macular Degeneration"]}],"id":[{"id":"https:\/\/ror.org\/03cvfxv40","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2025,9,30]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>Differentially active enhancers are key drivers of cell type specific gene expression. Active enhancers are found in open chromatin, which can be mapped at genome scale across tissue and cell types. Though incompletely understood, the relationship between chromatin accessibility and enhancer activity has been exploited to identify, model, and even design functional enhancers for selected cell types, but to what extent this design strategy can generalize across human cell and tissue types remains unclear. Here, we trained deep neural networks on a large corpus of chromatin accessibility data from hundreds of human biosamples. We used these models to generate an atlas of tens of thousands of synthetic enhancers, targeting hundreds of cell lines, tissues, and differentiation states, aiming to maximize accessibility in target samples and minimize it in all off-target ones. Experimental testing of thousands of designs in a representative subset of ten human cell types and in mouse retina demonstrated their function as specific enhancers, not only in the case of one-versus-all objectives but also when targeting two or three cell types. An explainable AI analysis, enabled by our large-scale enhancer measurements, allowed us to identify similarities and differences between the sequence grammar underlying accessibility and enhancer activity. Our results show that model-guided design of enhancers can help us decipher the cis-regulatory code governing cell type specificity and generate novel tools for selective targeting of human cell states.<\/jats:p>","DOI":"10.1101\/2025.09.30.679565","type":"posted-content","created":{"date-parts":[[2025,10,1]],"date-time":"2025-10-01T04:55:14Z","timestamp":1759294514000},"source":"Crossref","is-referenced-by-count":5,"title":["Programming human cell type-specific gene expression via an atlas of AI-designed enhancers"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2384-3129","authenticated-orcid":false,"given":"Sebastian M.","family":"Castillo-Hair","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christopher H.","family":"Yin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Leah","family":"VandenBosch","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Timothy J.","family":"Cherry","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wouter","family":"Meuleman","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Georg","family":"Seelig","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2025100320400655000_2025.09.30.679565v1.1","doi-asserted-by":"publisher","DOI":"10.7554\/eLife.27041"},{"key":"2025100320400655000_2025.09.30.679565v1.2","doi-asserted-by":"crossref","unstructured":"Dunbar, C. E. et al. Gene therapy comes of age. Science 359, eaan4672 (2018).","DOI":"10.1126\/science.aan4672"},{"key":"2025100320400655000_2025.09.30.679565v1.3","doi-asserted-by":"publisher","DOI":"10.1016\/j.copbio.2019.03.016"},{"key":"2025100320400655000_2025.09.30.679565v1.4","doi-asserted-by":"crossref","first-page":"41","DOI":"10.1016\/j.coisb.2019.02.008","article-title":"Synthetic development: learning to program multicellular self-organization","volume":"14","year":"2019","journal-title":"Curr. Opin. Syst. Biol"},{"key":"2025100320400655000_2025.09.30.679565v1.5","first-page":"1730","article-title":"Sensing and guiding cell-state transitions by using genetically encoded endoribonuclease-mediated microRNA sensors. Nat","volume":"8","year":"2024","journal-title":"Biomed. Eng"},{"key":"2025100320400655000_2025.09.30.679565v1.6","doi-asserted-by":"crossref","unstructured":"Yin, C. et al. Iterative deep learning design of human enhancers exploits condensed sequence grammar to achieve cell-type specificity. Cell Syst. 16, (2025).","DOI":"10.1016\/j.cels.2025.101302"},{"key":"2025100320400655000_2025.09.30.679565v1.7","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06936-2"},{"key":"2025100320400655000_2025.09.30.679565v1.8","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-024-08070-z"},{"key":"2025100320400655000_2025.09.30.679565v1.9","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06905-9"},{"key":"2025100320400655000_2025.09.30.679565v1.10","doi-asserted-by":"publisher","DOI":"10.1101\/2025.04.02.646812"},{"key":"2025100320400655000_2025.09.30.679565v1.11","doi-asserted-by":"publisher","DOI":"10.1126\/science.1205527"},{"key":"2025100320400655000_2025.09.30.679565v1.12","doi-asserted-by":"publisher","DOI":"10.1089\/nat.2018.0734"},{"key":"2025100320400655000_2025.09.30.679565v1.13","doi-asserted-by":"publisher","DOI":"10.1101\/2024.10.28.620728"},{"key":"2025100320400655000_2025.09.30.679565v1.14","doi-asserted-by":"publisher","DOI":"10.1101\/2024.12.31.630783"},{"key":"2025100320400655000_2025.09.30.679565v1.15","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2016.09.018"},{"key":"2025100320400655000_2025.09.30.679565v1.16","doi-asserted-by":"publisher","DOI":"10.1038\/nrm3949"},{"key":"2025100320400655000_2025.09.30.679565v1.17","doi-asserted-by":"crossref","unstructured":"Song, L. & Crawford, G. E . DNase-seq: A High-Resolution Technique for Mapping Active Gene Regulatory Elements across the Genome from Mammalian Cells. Cold Spring Harb. Protoc. 2010, pdb.prot5384 (2010).","DOI":"10.1101\/pdb.prot5384"},{"key":"2025100320400655000_2025.09.30.679565v1.18","doi-asserted-by":"publisher","DOI":"10.1038\/s41596-022-00692-9"},{"key":"2025100320400655000_2025.09.30.679565v1.19","doi-asserted-by":"publisher","DOI":"10.7554\/eLife.69479"},{"key":"2025100320400655000_2025.09.30.679565v1.20","doi-asserted-by":"crossref","unstructured":"Mich, J. K. et al. Functional enhancer elements drive subclass-selective expression from mouse to primate neocortex. Cell Rep. 34, (2021).","DOI":"10.1016\/j.celrep.2021.108754"},{"key":"2025100320400655000_2025.09.30.679565v1.21","doi-asserted-by":"publisher","DOI":"10.1101\/2024.06.10.597244"},{"key":"2025100320400655000_2025.09.30.679565v1.22","doi-asserted-by":"publisher","DOI":"10.1101\/2023.08.21.554146"},{"key":"2025100320400655000_2025.09.30.679565v1.23","doi-asserted-by":"publisher","DOI":"10.1038\/s41580-022-00464-z"},{"key":"2025100320400655000_2025.09.30.679565v1.24","doi-asserted-by":"publisher","DOI":"10.1016\/j.tig.2023.10.007"},{"key":"2025100320400655000_2025.09.30.679565v1.25","doi-asserted-by":"publisher","DOI":"10.1038\/nbt.2137"},{"key":"2025100320400655000_2025.09.30.679565v1.26","doi-asserted-by":"publisher","DOI":"10.1126\/science.1232542"},{"key":"2025100320400655000_2025.09.30.679565v1.27","doi-asserted-by":"publisher","DOI":"10.1038\/s41588-022-01278-7"},{"key":"2025100320400655000_2025.09.30.679565v1.28","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-024-02260-3"},{"key":"2025100320400655000_2025.09.30.679565v1.29","doi-asserted-by":"publisher","DOI":"10.1101\/2023.11.19.567742"},{"key":"2025100320400655000_2025.09.30.679565v1.30","doi-asserted-by":"crossref","unstructured":"Maslova, A. , et al. Deep learning of immune cell differentiation. Proc. Natl. Acad. Sci. 117, 25655\u2013 25666 (2020).","DOI":"10.1073\/pnas.2011795117"},{"key":"2025100320400655000_2025.09.30.679565v1.31","doi-asserted-by":"publisher","DOI":"10.1101\/gr.260851.120"},{"key":"2025100320400655000_2025.09.30.679565v1.32","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-021-01252-x"},{"key":"2025100320400655000_2025.09.30.679565v1.33","doi-asserted-by":"publisher","DOI":"10.1038\/S41588-021-00782-6"},{"key":"2025100320400655000_2025.09.30.679565v1.34","doi-asserted-by":"publisher","DOI":"10.1038\/S41586-021-04262-Z"},{"key":"2025100320400655000_2025.09.30.679565v1.35","doi-asserted-by":"crossref","unstructured":"Agarwal, V. & Kelley, D. R . The genetic and biochemical determinants of mRNA degradation rates in mammals. Genome Biol. 23, 245 (2022).","DOI":"10.1186\/s13059-022-02811-x"},{"key":"2025100320400655000_2025.09.30.679565v1.36","doi-asserted-by":"publisher","DOI":"10.1101\/2024.08.27.609971"},{"key":"2025100320400655000_2025.09.30.679565v1.37","doi-asserted-by":"publisher","DOI":"10.1038\/s41588-024-02053-6"},{"key":"2025100320400655000_2025.09.30.679565v1.38","doi-asserted-by":"publisher","DOI":"10.1101\/2024.10.09.617507"},{"key":"2025100320400655000_2025.09.30.679565v1.39","doi-asserted-by":"publisher","DOI":"10.1101\/2024.09.19.613754"},{"key":"2025100320400655000_2025.09.30.679565v1.40","doi-asserted-by":"publisher","DOI":"10.1101\/2024.12.25.630221"},{"key":"2025100320400655000_2025.09.30.679565v1.41","doi-asserted-by":"publisher","DOI":"10.1038\/s41587-025-02712-x"},{"key":"2025100320400655000_2025.09.30.679565v1.42","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2019.04.046"},{"key":"2025100320400655000_2025.09.30.679565v1.43","doi-asserted-by":"publisher","DOI":"10.1038\/s41587-019-0164-5"},{"key":"2025100320400655000_2025.09.30.679565v1.44","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1008982"},{"key":"2025100320400655000_2025.09.30.679565v1.45","doi-asserted-by":"crossref","unstructured":"Linder, J. , Koplik, S. E. , Kundaje, A. & Seelig, G . Deciphering the impact of genetic variation on human polyadenylation using APARENT2. Genome Biol. 23, 232 (2022).","DOI":"10.1186\/s13059-022-02799-4"},{"key":"2025100320400655000_2025.09.30.679565v1.46","doi-asserted-by":"publisher","DOI":"10.1101\/gad.351800.124"},{"key":"2025100320400655000_2025.09.30.679565v1.47","doi-asserted-by":"publisher","DOI":"10.1038\/s41576-025-00841-2"},{"key":"2025100320400655000_2025.09.30.679565v1.48","doi-asserted-by":"crossref","unstructured":"Castillo-Hair, S. et al. Optimizing 5\u2019UTRs for mRNA-delivered gene editing using deep learning. Nat. Commun. 15, 5284 (2024).","DOI":"10.1038\/s41467-024-49508-2"},{"key":"2025100320400655000_2025.09.30.679565v1.49","doi-asserted-by":"crossref","first-page":"374","DOI":"10.1038\/s44222-025-00280-y","article-title":"Modelling and design of transcriptional enhancers","volume":"3","year":"2025","journal-title":"Nat. Rev. Bioeng"},{"key":"2025100320400655000_2025.09.30.679565v1.50","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-020-2559-3"},{"key":"2025100320400655000_2025.09.30.679565v1.51","doi-asserted-by":"publisher","DOI":"10.1186\/s13059-024-03198-7"},{"key":"2025100320400655000_2025.09.30.679565v1.52","doi-asserted-by":"crossref","unstructured":"Linder, J. & Seelig, G . Fast activation maximization for molecular sequence design. BMC Bioinformatics 22, 510 (2021).","DOI":"10.1186\/s12859-021-04437-5"},{"key":"2025100320400655000_2025.09.30.679565v1.53","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1016\/j.cels.2020.05.007","article-title":"A Generative Neural Network for Maximizing Fitness and Diversity of Synthetic DNA and Protein Sequences","volume":"11","year":"2020","journal-title":"Cell Syst"},{"key":"2025100320400655000_2025.09.30.679565v1.54","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/bti042"},{"key":"2025100320400655000_2025.09.30.679565v1.55","unstructured":"Lundberg, S. M. & Lee, S.-I. A Unified Approach to Interpreting Model Predictions. in Advances in Neural Information Processing Systems vol. 30 (Curran Associates, Inc., 2017)."},{"key":"2025100320400655000_2025.09.30.679565v1.56","doi-asserted-by":"crossref","unstructured":"Lima Cunha, D. , Arno, G. , Corton, M. & Moosajee, M. The Spectrum of PAX6 Mutations and Genotype-Phenotype Correlations in the Eye. Genes 10, 1050 (2019).","DOI":"10.3390\/genes10121050"},{"key":"2025100320400655000_2025.09.30.679565v1.57","doi-asserted-by":"publisher","DOI":"10.1074\/jbc.M114.556738"},{"key":"2025100320400655000_2025.09.30.679565v1.58","doi-asserted-by":"publisher","DOI":"10.1128\/MCB.25.14.6031-6046.2005"},{"key":"2025100320400655000_2025.09.30.679565v1.59","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2012.08.027"},{"key":"2025100320400655000_2025.09.30.679565v1.60","doi-asserted-by":"publisher","DOI":"10.1161\/CIRCRESAHA.116.310456"},{"key":"2025100320400655000_2025.09.30.679565v1.61","doi-asserted-by":"publisher","DOI":"10.1016\/j.molcel.2017.11.026"},{"key":"2025100320400655000_2025.09.30.679565v1.62","doi-asserted-by":"publisher","DOI":"10.1016\/j.tcb.2018.04.009"},{"key":"2025100320400655000_2025.09.30.679565v1.63","doi-asserted-by":"crossref","unstructured":"Maytum, A. , Obier, N. , Cauchy, P. & Bonifer, C . Regulation of developmentally controlled enhancer activity by extrinsic signals in normal and malignant cells: AP-1 at the centre. Front. Epigenetics Epigenomics 2, (2024).","DOI":"10.3389\/freae.2024.1465958"},{"key":"2025100320400655000_2025.09.30.679565v1.64","doi-asserted-by":"publisher","DOI":"10.1101\/gr.279142.124"},{"key":"2025100320400655000_2025.09.30.679565v1.65","doi-asserted-by":"publisher","DOI":"10.1101\/2024.02.01.578352"},{"key":"2025100320400655000_2025.09.30.679565v1.66","doi-asserted-by":"crossref","unstructured":"Li, L. & Wunderlich, Z . An Enhancer\u2019s Length and Composition Are Shaped by Its Regulatory Task. Front. Genet. 8, (2017).","DOI":"10.3389\/fgene.2017.00063"},{"key":"2025100320400655000_2025.09.30.679565v1.67","doi-asserted-by":"publisher","DOI":"10.1038\/s41576-020-00311-x"},{"key":"2025100320400655000_2025.09.30.679565v1.68","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2015.01.006"},{"key":"2025100320400655000_2025.09.30.679565v1.69","doi-asserted-by":"crossref","unstructured":"Li, S. , Hannenhalli, S. & Ovcharenko, I . De novo human brain enhancers created by single-nucleotide mutations. Sci. Adv. 9, eadd2911 (2023).","DOI":"10.1126\/sciadv.add2911"},{"key":"2025100320400655000_2025.09.30.679565v1.70","doi-asserted-by":"publisher","DOI":"10.1126\/science.1246426"},{"key":"2025100320400655000_2025.09.30.679565v1.71","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06922-8"},{"key":"2025100320400655000_2025.09.30.679565v1.72","doi-asserted-by":"publisher","DOI":"10.1016\/j.devcel.2023.09.005"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1101\/2025.09.30.679565","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T09:42:00Z","timestamp":1768470120000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.1101\/2025.09.30.679565"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,30]]},"references-count":72,"URL":"https:\/\/doi.org\/10.1101\/2025.09.30.679565","relation":{},"subject":[],"published":{"date-parts":[[2025,9,30]]},"subtype":"preprint"}}