Cell2Text: Multimodal LLM for generating textual descriptions from single-cell RNA-Seq profiles

Μαρκογιαννάκης, Άρης; Markogiannakis, Aris

dc.contributor.author	Μαρκογιαννάκης, Άρης	el
dc.contributor.author	Markogiannakis, Aris	en
dc.date.accessioned	2026-04-24T08:58:51Z
dc.date.available	2026-04-24T08:58:51Z
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/64348
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.32042
dc.rights	Default License
dc.subject	Deep Learning	en
dc.subject	AI4Science	en
dc.subject	Multimodality	en
dc.subject	Natural Language Generation	en
dc.subject	Single-cell RNA-seq	en
dc.title	Cell2Text: Multimodal LLM for generating textual descriptions from single-cell RNA-Seq profiles	en
heal.type	bachelorThesis
heal.classification	Artificial Intelligence	en
heal.language	en
heal.access	free
heal.recordProvider	ntua	el
heal.publicationDate	2025-11-05
heal.abstract	Single-cell RNA sequencing has revolutionized biological research by enabling gene expression measurement at cellular resolution, revealing diverse cell types, states, and disease contexts. Recent single-cell foundation models can learn generalizable representations directly from expression data, improving downstream classification and clustering tasks. However, such models typically rely on fixed label spaces that limit their ability to express cellular diversity. This thesis presents Cell2Text, a multimodal generative framework that transforms single-cell transcriptomic profiles into structured natural language descriptions. By integrating pretrained single-cell encoders with large language models through learnable projection modules, Cell2Text generates coherent summaries describing cellular identity, tissue of origin, disease relevance, and biological pathway activity. Experimental results show that Cell2Text achieves higher accuracy than baseline models, maintains strong ontological consistency through PageRank-based similarity metrics, and produces semantically faithful text outputs. Overall, the proposed approach highlights the potential of combining biological and linguistic representations for scalable and informative single-cell characterization.	en
heal.advisorName	Stamou, Giorgos	en
heal.committeeMemberName	Vazirgiannis, Michalis	en
heal.committeeMemberName	Voulodimos, Athanasios	en
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Ηλεκτρολόγων Μηχανικών και Μηχανικών Υπολογιστών. Τομέας Τεχνολογίας Πληροφορικής και Υπολογιστών	el
heal.academicPublisherID	ntua
heal.numberOfPages	109 σ.	el
heal.fullTextAvailability	false