arXiv:2510.07793v3 Announce Type: replace 
Abstract: Large language models (LLMs) and emerging agentic frameworks are beginning to transform single-cell biology by enabling natural-language reasoning, generative annotation, and multimodal data integration. However, progress remains fragmented across data modalities, architectures, and evaluation standards. LLM4Cell presents the first unified survey of 58 foundation and agentic models developed for single-cell research, spanning RNA, ATAC, multi-omic, and spatial modalities. We categorize these methods into five families-foundation, text-bridge, spatial, multimodal, epigenomic, and agentic-and map them to eight key analytical tasks including annotation, trajectory and perturbation modeling, and drug-response prediction. Drawing on over 40 public datasets, we analyze benchmark suitability, data diversity, and ethical or scalability constraints, and evaluate models across 10 domain dimensions covering biological grounding, multi-omics alignment, fairness, privacy, and explainability. By linking datasets, models, and evaluation domains, LLM4Cell provides the first integrated view of language-driven single-cell intelligence and outlines open challenges in interpretability, standardization, and trustworthy model development.

ظهرت LLM4Cell كاستطلاع مهم لنماذج اللغة الكبيرة والنماذج الوكيلة المصممة لبيولوجيا الخلايا الفردية، حيث تحلل 58 نموذجًا عبر عدة أنماط، بما في ذلك RNA وATAC. تصنف هذه المراجعة الشاملة الأساليب إلى خمس عائلات وتربطها بثماني مهام تحليلية، مما يبرز التقدم المجزأ في هذا المجال.

LLM4Cell ha surgido como una encuesta significativa de grandes modelos de lenguaje y modelos agenticos diseñados para la biología de células individuales, analizando 58 modelos a través de diversas modalidades, incluyendo ARN y ATAC. Esta revisión integral categoriza los métodos en cinco familias y los mapea a ocho tareas analíticas, destacando el progreso fragmentado en el campo.

LLM4Cell se présente comme une enquête significative sur les grands modèles de langage et les modèles agentiques adaptés à la biologie des cellules uniques, analysant 58 modèles à travers diverses modalités, y compris l'ARN et l'ATAC. Cette revue complète catégorise les méthodes en cinq familles et les associe à huit tâches analytiques, mettant en lumière les progrès fragmentés dans le domaine.

LLM4Cell has emerged as a significant survey of large language and agentic models tailored for single-cell biology, analyzing 58 models across various modalities including RNA and ATAC. This comprehensive review categorizes methods into five families and maps them to eight analytical tasks, highlighting the fragmented progress in the field.

LLM4Cell: A Survey of Large Language and Agentic Models for Single-Cell Biology

طور الباحثون مفاتيح جزيئية من الحمض النووي الريبي باستخدام آلات بولتزمان المقيدة، كما ورد في مجلة Nature — Machine Learning. تهدف هذه الطريقة المبتكرة إلى تحسين تصميم ووظائف جزيئات الحمض النووي الريبي، التي تعتبر حيوية لعمليات بيولوجية متعددة.

Investigadores han desarrollado interruptores moleculares de ARN utilizando máquinas de Boltzmann restringidas, según se informa en Nature — Machine Learning. Este enfoque innovador tiene como objetivo mejorar el diseño y la funcionalidad de las moléculas de ARN, que son cruciales para varios procesos biológicos.

Des chercheurs ont développé des commutateurs moléculaires d'ARN en utilisant des machines de Boltzmann restreintes, comme rapporté dans Nature — Machine Learning. Cette approche innovante vise à améliorer la conception et la fonctionnalité des molécules d'ARN, qui sont cruciales pour divers processus biologiques.

Researchers have developed molecular RNA switches using Restricted Boltzmann machines, as reported in Nature — Machine Learning. This innovative approach aims to enhance the design and functionality of RNA molecules, which are crucial for various biological processes.

Designing molecular RNA switches with Restricted Boltzmann machines

One More Thing in AI – Your Shortcut to AI Mastery

LLM4Cell: A Survey of Large Language and Agentic Models for Single-Cell Biology

Was this article worth reading? Share it

One More Thing in AI

LucidQuery AI

Magicley AI

Supametas.AI

ChatOne

Legion AI

Ready to build your own newsroom?