Cell2Text: Multimodal LLM for Generating Single-Cell Descriptions from RNA-Seq Data

2025-09-29

AI Summary

Current single-cell foundation models produce only discrete labels, failing to deliver biologically rich, interpretable cell descriptions required by domain experts. To address this limitation, we propose the first generative framework that directly maps single-cell RNA-seq data to structured natural language descriptions—encompassing cell type, tissue origin, disease associations, and pathway activity. Our method employs a multimodal learning architecture that jointly leverages gene-level embeddings from a single-cell foundation model and a pretrained large language model, enabling end-to-end sequence-to-text generation. Experiments demonstrate that our framework outperforms classification baselines in label accuracy while achieving significant improvements in semantic fidelity (e.g., BERTScore) and ontology-aware similarity (e.g., Cell Ontology-based path similarity). Crucially, it delivers high interpretability through human-readable outputs and strong generalization across diverse datasets and cell types.

Technology Category

• Natural Language Processing: Generation
• Machine Learning: Multimodal Learning
• Computer Vision: Language and Vision

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Single-cell RNA sequencing has transformed biology by enabling the measurement of gene expression at cellular resolution, providing information for cell types, states, and disease contexts. Recently, single-cell foundation models have emerged as powerful tools for learning transferable representations directly from expression profiles, improving performance on classification and clustering tasks. However, these models are limited to discrete prediction heads, which collapse cellular complexity into predefined labels that fail to capture the richer, contextual explanations biologists need. We introduce Cell2Text, a multimodal generative framework that translates scRNA-seq profiles into structured natural language descriptions. By integrating gene-level embeddings from single-cell foundation models with pretrained large language models, Cell2Text generates coherent summaries that capture cellular identity, tissue origin, disease associations, and pathway activity, generalizing to unseen cells. Empirically, Cell2Text outperforms baselines on classification accuracy, demonstrates strong ontological consistency using PageRank-based similarity metrics, and achieves high semantic fidelity in text generation. These results demonstrate that coupling expression data with natural language offers both stronger predictive performance and inherently interpretable outputs, pointing to a scalable path for label-efficient characterization of unseen cells.

Problem

• Generates natural language descriptions from single-cell RNA data
• Translates gene expression profiles into structured cellular summaries
• Overcomes limitations of discrete classification labels in biology

Innovation

• Generates natural language descriptions from RNA-seq data
• Integrates gene embeddings with large language models
• Produces interpretable summaries of cellular characteristics