Complete spatially resolved gene expression is not necessary for identifying spatial domains

Senlin Lin; Yan Cui; Fangyuan Zhao; Zhidong Yang; Jiangning Song; Jianhua Yao; Yu Zhao; Bin-Zhi Qian; Yi Zhao; Zhiyuan Yuan

doi:10.1016/j.xgen.2024.100565

Complete spatially resolved gene expression is not necessary for identifying spatial domains

Cell Genom. 2024 Jun 12;4(6):100565. doi: 10.1016/j.xgen.2024.100565. Epub 2024 May 22.

Authors

Senlin Lin¹, Yan Cui², Fangyuan Zhao¹, Zhidong Yang¹, Jiangning Song³, Jianhua Yao⁴, Yu Zhao⁴, Bin-Zhi Qian⁵, Yi Zhao⁶, Zhiyuan Yuan⁷

Affiliations

¹ Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China; University of Chinese Academy of Sciences, Beijing, China.
² Institute of Science and Technology for Brain-Inspired Intelligence, MOE Key Laboratory of Computational Neuroscience and Brain-Inspired Intelligence, MOE Frontiers Center for Brain Science, Fudan University, Shanghai, China; Center for Medical Research and Innovation, Shanghai Pudong Hospital, Fudan University Pudong Medical Center, Fudan University, Shanghai, China.
³ Biomedicine Discovery Institute and Department of Biochemistry and Molecular Biology, Monash University, Clayton, VIC 3800, Australia.
⁴ AI Lab, Tencent, Shenzhen, China.
⁵ Fudan University Shanghai Cancer Center, Department of Oncology, Shanghai Medical College, The Human Phenome Institute, Zhangjiang-Fudan International Innovation Center, Fudan University, Shanghai, China.
⁶ Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China; University of Chinese Academy of Sciences, Beijing, China. Electronic address: [email protected].
⁷ Institute of Science and Technology for Brain-Inspired Intelligence, MOE Key Laboratory of Computational Neuroscience and Brain-Inspired Intelligence, MOE Frontiers Center for Brain Science, Fudan University, Shanghai, China; Center for Medical Research and Innovation, Shanghai Pudong Hospital, Fudan University Pudong Medical Center, Fudan University, Shanghai, China. Electronic address: [email protected].

Abstract

Spatially resolved transcriptomics (SRT) technologies have revolutionized the study of tissue organization. We introduce a graph convolutional network with an attention and positive emphasis mechanism, termed BINARY, relying exclusively on binarized SRT data to accurately delineate spatial domains. BINARY outperforms existing methods across various SRT data types while using significantly less input information. Our study suggests that precise gene expression quantification may not always be essential, inspiring further exploration of the broader applications of spatially resolved binarized gene expression data.

Keywords: binary spatially resolved gene expression; bioinformatics; computational biology; graph convolutional network; spatial clustering; spatial domain; spatial omics; spatially resolved gene expression; spatially resolved transcriptomics; spatially variable gene.

MeSH terms

Algorithms
Gene Expression Profiling* / methods
Humans
Transcriptome / genetics