Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zhang, Heming, Xu, Tim, Cao, Dekang, Liang, Shunning, Shergill, Guntaas, Hadas, Nicholas, Schimmelpfennig, Lars, Kaster, Levi, Huang, Di, Li, Guangfu, Goedegebuure, S. Peter, DeNardo, David, Ding, Li, Fields, Ryan C., Miller, J Philip, Eghtesady, Pirooz, Cruchaga, Carlos, Buchser, William, Cooper, Jonathan, Sardiello, Marco, Dickson, Patricia, Chen, Yixin, Province, Michael, Payne, Philip, Li, Fuhai
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2504.02148
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866911420116893696
author Zhang, Heming
Xu, Tim
Cao, Dekang
Liang, Shunning
Shergill, Guntaas
Hadas, Nicholas
Schimmelpfennig, Lars
Kaster, Levi
Huang, Di
Li, Guangfu
Goedegebuure, S. Peter
DeNardo, David
Ding, Li
Fields, Ryan C.
Miller, J Philip
Eghtesady, Pirooz
Cruchaga, Carlos
Buchser, William
Cooper, Jonathan
Sardiello, Marco
Dickson, Patricia
Chen, Yixin
Province, Michael
Payne, Philip
Li, Fuhai
author_facet Zhang, Heming
Xu, Tim
Cao, Dekang
Liang, Shunning
Shergill, Guntaas
Hadas, Nicholas
Schimmelpfennig, Lars
Kaster, Levi
Huang, Di
Li, Guangfu
Goedegebuure, S. Peter
DeNardo, David
Ding, Li
Fields, Ryan C.
Miller, J Philip
Eghtesady, Pirooz
Cruchaga, Carlos
Buchser, William
Cooper, Jonathan
Sardiello, Marco
Dickson, Patricia
Chen, Yixin
Province, Michael
Payne, Philip
Li, Fuhai
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2504_02148
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle OmniCellTOSG: The First Cell Text-Omic Signaling Graphs Dataset for Graph Language Foundation Modeling
Zhang, Heming
Xu, Tim
Cao, Dekang
Liang, Shunning
Shergill, Guntaas
Hadas, Nicholas
Schimmelpfennig, Lars
Kaster, Levi
Huang, Di
Li, Guangfu
Goedegebuure, S. Peter
DeNardo, David
Ding, Li
Fields, Ryan C.
Miller, J Philip
Eghtesady, Pirooz
Cruchaga, Carlos
Buchser, William
Cooper, Jonathan
Sardiello, Marco
Dickson, Patricia
Chen, Yixin
Province, Michael
Payne, Philip
Li, Fuhai
Artificial Intelligence
Machine Learning
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.
title OmniCellTOSG: The First Cell Text-Omic Signaling Graphs Dataset for Graph Language Foundation Modeling
topic Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2504.02148