Cell-GraphCompass: modeling single cells with graph structure foundation model

doi:10.1093/nsr/nwaf255

	Cell-GraphCompass: modeling single cells with graph structure foundation model
	Fang, Chen 1,2,3,4,5; Cui, Wentao 4,6; Hu, Zhilong 4,5,6; Liu, Wenhao 2,3; Chen, Shubai 6,7; Chang, Shaole 5; Long, Qingqing 4,6; Li, Cong 2,3,6; Liu, Yana 2,3; Jiang, Haiping 2,3; Wang, Pengfei 4; Pan, Jia 5; Hu, Guoping 5; Liu, Guole 2,3,6,8; Meng, Zhen 4,6; Zhou, Yuanchun 1,4; Chen, Linghui 9; Feng, Guihai 2,3,6,8; Li, Xin 1,2,3,8
	2025-10-01
发表期刊	NATIONAL SCIENCE REVIEW
ISSN	2095-5138
卷号	12 期号:10 页码:14
摘要	Cells in the human body are regulated by sophisticated networks of gene regulation, which allows them to fulfill their cellular destiny and function. Inspired by the advancements in large language models, there have been several attempts focusing on constructing foundation models with single-cell transcriptomic data to decipher gene regulatory networks. However, these models tend to impose a sequential structure on genes within each cell, which may omit intrinsic biological characteristics and lack the utilization of other available prior knowledge. In this paper, we introduce Cell-GraphCompass (CGCompass), the pioneering foundation model that employs graph pre-training to model genes and cells. We use three types of gene-related information as node features for constructing cell graphs and collect data from three perspectives depicting relationships between genes as edge features. We pre-trained the model with over 50 million human cells and then fine-tuned it to a broad spectrum of tasks, such as batch integration, cell type annotation, single-cell gene perturbation and in silico gene knockout predictions, achieving commendable performance. Overall, CGCompass provides a practical architecture for leveraging graph pre-training to incorporate prior knowledge in constructing a foundation model for single-cell analysis.
关键词	single-cell transcriptomics foundation model pre-training graph neural network knowledge embedding
DOI	10.1093/nsr/nwaf255
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Program ; Computer Network Information Center, Chinese Academy of Sciences
WOS研究方向	Science & Technology - Other Topics
WOS类目	Multidisciplinary Sciences
WOS记录号	WOS:001584636200001
出版者	OXFORD UNIV PRESS
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/41645
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Meng, Zhen; Zhou, Yuanchun; Chen, Linghui; Feng, Guihai; Li, Xin
作者单位	1.Univ Chinese Acad Sci, Sch Adv Interdisciplinary Sci, Beijing 100049, Peoples R China 2.Chinese Acad Sci, Inst Zool, State Key Lab Organ Regenerat & Reconstruct, Beijing 100101, Peoples R China 3.Beijing Inst Stem Cell & Regenerat Med, Beijing 100101, Peoples R China 4.Chinese Acad Sci, Comp Network Informat Ctr, Beijing 100083, Peoples R China 5.iFLYTEK Res, Hefei 230088, Peoples R China 6.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 7.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China 8.Chinese Acad Sci, Inst Zool, Human Organ Physiopathol Emulat Syst, Beijing 100101, Peoples R China 9.Oristruct Biotech Co Ltd, Hefei 230026, Peoples R China
推荐引用方式 GB/T 7714	Fang, Chen,Cui, Wentao,Hu, Zhilong,et al. Cell-GraphCompass: modeling single cells with graph structure foundation model[J]. NATIONAL SCIENCE REVIEW,2025,12(10):14.
APA	Fang, Chen.,Cui, Wentao.,Hu, Zhilong.,Liu, Wenhao.,Chen, Shubai.,...&Li, Xin.(2025).Cell-GraphCompass: modeling single cells with graph structure foundation model.NATIONAL SCIENCE REVIEW,12(10),14.
MLA	Fang, Chen,et al."Cell-GraphCompass: modeling single cells with graph structure foundation model".NATIONAL SCIENCE REVIEW 12.10(2025):14.