{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,10]],"date-time":"2026-09-10T20:09:10Z","timestamp":1789070950457,"version":"build-2803163510"},"reference-count":38,"publisher":"Oxford University Press (OUP)","issue":"5","license":[{"start":{"date-parts":[[2026,9,10]],"date-time":"2026-09-10T00:00:00Z","timestamp":1788998400000},"content-version":"vor","delay-in-days":9,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Dr Antonio Orvieto"},{"name":"Max Planck Institute for Intelligent Systems"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,9,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Translating single-cell RNA sequencing (scRNA-seq) data into mechanistic biological hypotheses remains a critical bottleneck, as agentic AI systems lack direct access to transcriptomic representations while expression foundation models remain opaque to natural language. Here, we introduce ELISA (Embedding-Linked Interactive Single-cell Agent), an interpretable framework that unifies single-cell generative pretrained transformer expression embeddings with biomedical bidirectional encoder representations from transformers-based semantic retrieval and large-language model (LLM)-mediated interpretation for interactive single-cell discovery. An automatic query classifier routes inputs to gene marker scoring, semantic matching, or reciprocal rank fusion pipelines depending on whether the query is a gene signature, natural language concept, or mixture of both. Integrated analytical modules perform pathway activity scoring across 60+ gene sets, ligand\u2013receptor interaction prediction using 280+ curated pairs, condition-aware comparative analysis, and cell-type proportion estimation, all operating directly on embedded data without access to the original count matrix. Benchmarked across six diverse scRNA-seq datasets spanning inflammatory lung disease, pediatric and adult cancers, organoid models, healthy tissue, and neurodevelopment, ELISA significantly outperforms CellWhisperer, a classical lexical retriever (BM25), and a random baseline in cell type retrieval (combined permutation test, $p &amp;lt; 2\\times 10^{-5}$ for each), with particularly large gains on gene-signature queries (Cohen\u2019s $d = 5.98$ for mean reciprocal rank). ELISA replicates published biological findings (mean composite score 0.88), and generates candidate hypotheses through grounded LLM reasoning, bridging the gap between transcriptomic data exploration and biological discovery.<\/jats:p>","DOI":"10.1093\/bib\/bbag501","type":"journal-article","created":{"date-parts":[[2026,9,4]],"date-time":"2026-09-04T11:19:45Z","timestamp":1788520785000},"source":"Crossref","is-referenced-by-count":0,"title":["ELISA (Embedding-Linked Interactive Single-cell Agent): an interpretable hybrid generative Artificial Intelligence agent for expression-grounded discovery in single-cell genomics"],"prefix":"10.1093","volume":"27","author":[{"given":"Omar","family":"Coser","sequence":"first","affiliation":[{"name":"Independent Researcher , Via Vecchie Pontare 24, 38121 Meano, Trento, Italy, Europe"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,9,10]]},"reference":[{"key":"2026091015383307100_ref1","doi-asserted-by":"crossref","first-page":"377","DOI":"10.1038\/nmeth.1315","article-title":"mRNA-seq whole-transcriptome analysis of a single cell","volume":"6","author":"Tang","year":"2009","journal-title":"Nat Methods"},{"key":"2026091015383307100_ref2","doi-asserted-by":"crossref","first-page":"e8746","DOI":"10.15252\/msb.20188746","article-title":"Current best practices in single-cell RNA-seq analysis: a tutorial","volume":"15","author":"Luecken","year":"2019","journal-title":"Mol Syst Biol"},{"key":"2026091015383307100_ref3","doi-asserted-by":"crossref","first-page":"172","DOI":"10.1038\/s41586-023-06291-2","article-title":"Large language models encode clinical knowledge","volume":"620","author":"Singhal","year":"2023","journal-title":"Nature"},{"key":"2026091015383307100_ref4","first-page":"9459","article-title":"Retrieval-augmented generation for knowledge-intensive NLP tasks","volume":"33","author":"Lewis","year":"2020","journal-title":"Adv Neural Inf Proces Syst"},{"key":"2026091015383307100_ref5","doi-asserted-by":"crossref","DOI":"10.1038\/s41586-026-10644-y","article-title":"Accelerating scientific discovery with Co-Scientist","volume":"655","author":"Gottweis","year":"2026","journal-title":"Nature"},{"key":"2026091015383307100_ref6","doi-asserted-by":"crossref","DOI":"10.1101\/2025.05.30.656746","article-title":"Biomni: a general-purpose biomedical AI agent","author":"Huang","year":"2025"},{"key":"2026091015383307100_ref7","doi-asserted-by":"crossref","first-page":"1677","DOI":"10.1038\/s41592-025-02748-6","article-title":"Geneagent: self-verification language agent for gene-set analysis using domain databases","volume":"22","author":"Wang","year":"2025","journal-title":"Nat Methods"},{"key":"2026091015383307100_ref8","doi-asserted-by":"crossref","first-page":"6125","DOI":"10.1016\/j.cell.2024.09.022","article-title":"Empowering biomedical discovery with AI agents","volume":"187","author":"Gao","year":"2024","journal-title":"Cell"},{"key":"2026091015383307100_ref9","doi-asserted-by":"crossref","first-page":"716","DOI":"10.1038\/s41586-025-09442-9","article-title":"The virtual lab of AI agents designs new SARS-CoV-2 nanobodies","volume":"646","author":"Swanson","year":"2025","journal-title":"Nature"},{"key":"2026091015383307100_ref10","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2407.09811","article-title":"CellAgent: an LLM-driven multi-agent framework for automated single-cell data analysis","author":"Xiao","year":"2024"},{"key":"2026091015383307100_ref11","doi-asserted-by":"crossref","DOI":"10.1002\/advs.202407094","article-title":"An AI agent for fully automated multi-omic analyses","volume":"11","author":"Zhou","year":"2024","journal-title":"Advanced Science"},{"key":"2026091015383307100_ref12","doi-asserted-by":"crossref","first-page":"btaf159","DOI":"10.1093\/bioinformatics\/btaf159","article-title":"Automatic biomarker discovery and enrichment with brad","volume":"41","author":"Pickard","year":"2025","journal-title":"Bioinformatics"},{"key":"2026091015383307100_ref13","doi-asserted-by":"crossref","first-page":"btae075","DOI":"10.1093\/bioinformatics\/btae075","article-title":"GeneGPT: augmenting large language models with domain tools for improved access to biomedical information","volume":"40","author":"Jin","year":"2024","journal-title":"Bioinformatics"},{"key":"2026091015383307100_ref14","doi-asserted-by":"crossref","first-page":"91","DOI":"10.1186\/s13073-025-01521-w","article-title":"Improving automated deep phenotyping through large language models using retrieval-augmented generation","volume":"17","author":"Garcia","year":"2025","journal-title":"Genome Med"},{"key":"2026091015383307100_ref15","doi-asserted-by":"crossref","first-page":"bbad535","DOI":"10.1093\/bib\/bbad535","article-title":"ChIP-GPT: a managed large language model for robust data extraction from biomedical database records","volume":"25","author":"Cinquin","year":"2024","journal-title":"Brief Bioinform"},{"key":"2026091015383307100_ref16","doi-asserted-by":"crossref","first-page":"vbaf141","DOI":"10.1093\/bioadv\/vbaf141","article-title":"Leveraging generative AI to assist biocuration of medical actions for rare disease","volume":"5","author":"Enock Niyonkuru","year":"2025","journal-title":"Bioinform Adv"},{"key":"2026091015383307100_ref17","first-page":"245","article-title":"CRISPR-GPT for agentic automation of gene-editing experiments","volume":"10","author":"Yuanhao","year":"2026","journal-title":"NatBiomed Eng"},{"key":"2026091015383307100_ref18","doi-asserted-by":"crossref","first-page":"1470","DOI":"10.1038\/s41592-024-02201-0","article-title":"scGPT: toward building a foundation model for single-cell multi-omics using generative AI","volume":"21","author":"Cui","year":"2024","journal-title":"Nat Methods"},{"key":"2026091015383307100_ref19","doi-asserted-by":"publisher","DOI":"10.1101\/2025.02.19.638959","article-title":"scWGBS-GPT: a foundation model for capturing long-range CpG dependencies in single-cell whole-genome bisulfite sequencing to enhance epigenetic analysis","author":"Liang","year":"2025"},{"key":"2026091015383307100_ref20","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2403.11073","article-title":"Tokensome: towards a genetic vision-language GPT for explainable and cognitive karyotyping","author":"Zhang","year":"2024"},{"key":"2026091015383307100_ref21","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41587-025-02857-9","article-title":"Multimodal learning enables chat-based exploration of single-cell data","author":"Schaefer","year":"2025","journal-title":"Nat Biotechnol"},{"key":"2026091015383307100_ref22","doi-asserted-by":"crossref","DOI":"10.1016\/j.jcf.2025.01.016","article-title":"Evidence for altered immune-structural cell crosstalk in cystic fibrosis revealed by single cell transcriptomics","volume":"24","author":"Berg","year":"2025","journal-title":"J Cyst Fibros"},{"key":"2026091015383307100_ref23","doi-asserted-by":"crossref","first-page":"1142","DOI":"10.1038\/s41588-025-02158-6","article-title":"Longitudinal single-cell multiomic atlas of high-risk neuroblastoma reveals chemotherapy-induced tumor microenvironment rewiring","volume":"57","author":"Wenbao","year":"2025","journal-title":"Nat Genet"},{"key":"2026091015383307100_ref24","doi-asserted-by":"crossref","first-page":"139","DOI":"10.1038\/s41597-025-04381-6","article-title":"Integrated cancer cell-specific single-cell RNA-seq datasets of immune checkpoint blockade-treated patients","volume":"12","author":"Gondal","year":"2025","journal-title":"Sci Data"},{"key":"2026091015383307100_ref25","doi-asserted-by":"crossref","first-page":"639","DOI":"10.1038\/s44318-024-00328-6","article-title":"A novel human fetal lung-derived alveolar organoid model reveals mechanisms of surfactant protein C maturation relevant to interstitial lung disease","volume":"44","author":"Lim","year":"2025","journal-title":"EMBO J"},{"key":"2026091015383307100_ref26","doi-asserted-by":"crossref","first-page":"3482","DOI":"10.1038\/s41591-024-03011-9","article-title":"Single-nucleus chromatin accessibility and transcriptomic map of breast tissues of women of diverse genetic ancestry","volume":"30","author":"Bhat-Nakshatri","year":"2024","journal-title":"Nat Med"},{"key":"2026091015383307100_ref27","doi-asserted-by":"crossref","first-page":"179","DOI":"10.1038\/s41586-024-07234-1","article-title":"Chromatin accessibility during human first-trimester neurodevelopment","volume":"647","author":"Mannens","year":"2025","journal-title":"Nature"},{"key":"2026091015383307100_ref28","doi-asserted-by":"crossref","first-page":"1234","DOI":"10.1093\/bioinformatics\/btz682","article-title":"BioBERT: a pre-trained biomedical language representation model for biomedical text mining","volume":"36","author":"Lee","year":"2020","journal-title":"Bioinformatics"},{"key":"2026091015383307100_ref29","doi-asserted-by":"crossref","first-page":"180","DOI":"10.1038\/s41596-024-01045-4","article-title":"CellChat for systematic analysis of cell\u2013cell communication from single-cell transcriptomics","volume":"20","author":"Jin","year":"2025","journal-title":"Nat Protoc"},{"key":"2026091015383307100_ref30","doi-asserted-by":"crossref","first-page":"1484","DOI":"10.1038\/s41596-020-0292-x","article-title":"CellPhoneDB: inferring cell\u2013cell communication from combined expression of multi-subunit ligand\u2013receptor complexes","volume":"15","author":"Efremova","year":"2020","journal-title":"Nat Protoc"},{"key":"2026091015383307100_ref31","doi-asserted-by":"crossref","first-page":"159","DOI":"10.1038\/s41592-019-0667-5","article-title":"NicheNet: modeling intercellular communication by linking ligands to target genes","volume":"17","author":"Browaeys","year":"2020","journal-title":"Nat Methods"},{"key":"2026091015383307100_ref32","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2407.21783","article-title":"The llama 3 herd of models","author":"Grattafiori","year":"2024"},{"key":"2026091015383307100_ref33","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2303.08774","article-title":"GPT-4 technical report","author":"Achiam","year":"2023"},{"key":"2026091015383307100_ref34","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2312.11805","article-title":"Gemini: a family of highly capable multimodal models","author":"Team","year":"2023"},{"key":"2026091015383307100_ref35","author":"Anthropic","year":"2024"},{"key":"2026091015383307100_ref36","author":"Robertson","year":"2009"},{"key":"2026091015383307100_ref37","doi-asserted-by":"crossref","DOI":"10.4324\/9780203771587","volume-title":"Statistical Power Analysis for the Behavioral Sciences","author":"Cohen","year":"2013"},{"key":"2026091015383307100_ref38","doi-asserted-by":"crossref","DOI":"10.1201\/9781003456285","volume-title":"Statistical Inference","author":"Casella","year":"2024"}],"container-title":["Briefings in Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/27\/5\/bbag501\/71022324\/bbag501.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/27\/5\/bbag501\/71022324\/bbag501.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,10]],"date-time":"2026-09-10T19:38:41Z","timestamp":1789069121000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bib\/article\/doi\/10.1093\/bib\/bbag501\/8790175"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,1]]},"references-count":38,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2026,9,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bib\/bbag501","relation":{},"ISSN":["1467-5463","1477-4054"],"issn-type":[{"value":"1467-5463","type":"print"},{"value":"1477-4054","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,9]]},"published":{"date-parts":[[2026,9,1]]},"article-number":"bbag501"}}