GeneCompass: deciphering universal gene regulatory mechanisms with a knowledge-informed cross-species foundation model

Deciphering universal gene regulatory mechanisms in diverse organisms holds great potential for advancing our knowledge of fundamental life processes and facilitating clinical applications. However, the traditional research paradigm primarily focuses on individual model organisms and does not integrate various cell types across species. Recent breakthroughs in single-cell sequencing and deep learning techniques present an unprecedented opportunity to address this challenge. In this study, we built an extensive dataset of over 120 million human and mouse single-cell transcriptomes. After data preprocessing, we obtained 101,768,420 single-cell transcriptomes and developed a knowledge-informed cross-species foundation model, named GeneCompass. During pre-training, GeneCompass effectively integrated four types of prior biological knowledge to enhance our understanding of gene regulatory mechanisms in a self-supervised manner. By fine-tuning for multiple downstream tasks, GeneCompass outperformed state-of-the-art models in diverse applications for a single species and unlocked new realms of cross-species biological investigations. We also employed GeneCompass to search for key factors associated with cell fate transition and showed that the predicted candidate genes could successfully induce the differentiation of human embryonic stem cells into the gonadal fate. Overall, GeneCompass demonstrates the advantages of using artificial intelligence technology to decipher universal gene regulatory mechanisms and shows tremendous potential for accelerating the discovery of critical cell fate regulators and candidate drug targets.

Authors

Institutions

Publication Details

Journal
Cell Research
Published
2024-10-07
DOI
https://doi.org/10.1038/s41422-024-01034-y
Citations
86
Primary Topic
Single-cell and spatial transcriptomics
Type
article
Field-Weighted Citation Impact
20.11

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

GeneCompass: deciphering universal gene regulatory mechanisms with a knowledge-informed cross-species foundation model

Yi Zhao, Yu Zou, Yao Tian, Shan Zong et al.
86 citations
Cell Research
Single-cell and spatial transcriptomics
20.11
article

GeneCompass: deciphering universal gene regulatory mechanisms with a knowledge-informed cross-species foundation model

Yi Zhao, Yu Zou, Yao Tian, Shan Zong, Yiqiang Chen, Wenhui Wu, Yiming Huang, Zhen Meng, Xiaoning Qi, Shubai Chen, Qingqing Long, Jiaxin Qin, Shirui Li, Huimin He, Zaitian Wang, Ran Zhang, Xuezhi Wang, Zhengting Zou, Hefan Miao, Ao Li, Chenhao Li, Haoyuan Wang, Xiaodong Yang, Yiyun Li, Qunlun Shen, Weiwei Zhai, Guole Liu, Chengrui Wang, Jiaheng Zhou, Guihai Feng, Yana Liu, Jingtao Guo, Baoyang Hu, Yilong Hu, Liqun Zhong, Qi Gu, Wuliang Huang, Yuanchun Zhou, Yaning Li, Meng Xiao, Fang Chen, Yixiao Zhao, Xinxin Lin, Longfei Han, Shihua Zhang, Lina Cui, Zhilong Hu, Jingxi Dong, Dechao Bu, Fei Gao, Xin Li, Hongmei Wang, Zheng Li, Wenhao Liu, Jiahao Zhang, Leqian Yu, Ping Xu, Ziwen Liu, Fei Li, Yiyang Zhang, Zichen Wang, Zhaohui Yang, Shuyang Jiang, Fan Li, Huanhuan Wu, Handong Li, Xingjian He, Zeyuan Zhang, Wentao Cui, Haiping Jiang, Yaoru Luo, Qicheng Chen, Yang Ge, Zijian Wang, Zixu Deng, Shuyu Guo, Chunyang Zhang, Qinxuan Luo, Yongshun Ren, Qinmeng Yang, Zhongming Liang, Chaofan He, Xin Qin, Kangning Dong, Yangang Wang, Zhenpeng Man, Jing Liu, Teng Wang, Zhufeng Xu, Jie Jiang, Jingqi Yu, Yike Xin, Cong Li, Wei Li, Kai Ma, Yong Wang, Yining Wang, Jiajia Wang, Pengfei Wang, Yan Chen
article en
86 citations

Abstract

Deciphering universal gene regulatory mechanisms in diverse organisms holds great potential for advancing our knowledge of fundamental life processes and facilitating clinical applications. However, the traditional research paradigm primarily focuses on individual model organisms and does not integrate various cell types across species. Recent breakthroughs in single-cell sequencing and deep learning techniques present an unprecedented opportunity to address this challenge. In this study, we built an extensive dataset of over 120 million human and mouse single-cell transcriptomes. After data preprocessing, we obtained 101,768,420 single-cell transcriptomes and developed a knowledge-informed cross-species foundation model, named GeneCompass. During pre-training, GeneCompass effectively integrated four types of prior biological knowledge to enhance our understanding of gene regulatory mechanisms in a self-supervised manner. By fine-tuning for multiple downstream tasks, GeneCompass outperformed state-of-the-art models in diverse applications for a single species and unlocked new realms of cross-species biological investigations. We also employed GeneCompass to search for key factors associated with cell fate transition and showed that the predicted candidate genes could successfully induce the differentiation of human embryonic stem cells into the gonadal fate. Overall, GeneCompass demonstrates the advantages of using artificial intelligence technology to decipher universal gene regulatory mechanisms and shows tremendous potential for accelerating the discovery of critical cell fate regulators and candidate drug targets.

Cell ResearchVol. 34(12)
Chinese Academy of Sciences (CN), Institute for Stem Cell Biology and Regenerative Medicine (IN), Institute of Computing Technology (CN), Shandong Institute of Automation (CN), National Center for Mathematics and Interdisciplinary Sciences (CN), Beijing Academy of Artificial Intelligence (CN), Computer Network Information Center (CN), Institute of Automation (CN), Academy of Mathematics and Systems Science (CN), University of Chinese Academy of Sciences (CN), Institute of Zoology (CN)
National Natural Science Foundation of China, Chinese Academy of Sciences, Institute of Zoology, Chinese Academy of Sciences
Life in Land
Openalex Percentile: Top 0%
Single-cell and spatial transcriptomics
20.11
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.