OmniCellTOSG: The First Cell Text-Omic Signaling Graphs Dataset for Graph Language Foundation Modeling

TL;DR

This paper introduces OmniCellTOSG, a large-scale, multimodal graph dataset integrating biomedical text, signaling networks, and omic data, enabling advanced, interpretable analysis of single-cell RNA sequencing across diseases.

Contribution

It presents the novel Text-Omic Signaling Graph (TOSG) data structure and constructs the OmniCellTOSG dataset, along with a new graph language foundation model for comprehensive single-cell data analysis.

Findings

CellTOSG-FM outperforms existing models on multiple tasks.

Provides interpretable insights into disease mechanisms.

Enables integration of textual, signaling, and omic data.

Abstract

With the rapid growth of large-scale single-cell omic datasets, omic foundation models (FMs) have emerged as powerful tools for advancing research in life sciences and precision medicine. However, most existing omic FMs rely primarily on numerical transcriptomic data by sorting genes as sequences, while lacking explicit integration of biomedical prior knowledge and signaling interactions that are critical for scientific discovery. Here, we introduce the Text-Omic Signaling Graph (TOSG), a novel data structure that unifies human-interpretable biomedical textual knowledge, quantitative omic data, and signaling network information. Using this framework, we construct OmniCellTOSG, a large-scale resource comprising approximately half million meta-cell TOSGs derived from around 80 million single-cell and single-nucleus RNA-seq profiles across organs and diseases. We further develop CellTOSG-FM, a multimodal graph language FM, to jointly analyze textual, omic and signaling network context. Across diverse downstream tasks, CellTOSG-FM outperforms existing omic FMs, and provides interpretable insights into disease-associated targets and signaling pathways.