EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Biosemantic graph transfer learning for cross-patient cell type annotation and cell–cell interaction analysis in tumor single-cell RNA sequencing

Yue-Chao Li iD, Leon Wong, Hai-Ru You, Xin-Fei Wang iD, Meng-Meng Wei iD, Lei Wang iD, Zhi-An Huang iD, Yu-An Huang iD, Zhu-Hong You iD

Briefings in Bioinformatics · 02.11.2026 · 27 (6)

DOI: 10.1093/bib/bbag587 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, online erschienen am 13.11.2026. Verfasst von Yue-Chao Li und 8 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science, Northwestern Polytechnical University, No.1 Dongxiang Road, Chang'an District, Xi'an 710129, Shaanxi, China. Hauptthemen laut MEDLINE: Cell Communication, Single-Cell Analysis, Sequence Analysis, RNA, Neoplasms. Die Arbeit nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Guangdong Basic and Applied Basic Research Foundation, Fundamental Research Funds for the Central Universities, Neural Science Foundation of Shaanxi Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The exploration of cellular heterogeneity within the tumor microenvironment (TME) via single-cell RNA sequencing (scRNA-seq) is essential for understanding cancer progression and response to therapy. Current scRNA-seq approaches, however, lack spatial context and rely on incomplete datasets of ligand–receptor interactions, limiting accurate cell type annotation and cell–cell communication inference. This study addresses these challenges using a novel graph neural network model that enhances cell-type prediction and cell-interaction analysis. Our study utilized a dataset consisting of 49 024 cells from 19 patients across three cancer types: leukemia, breast invasive carcinoma, and colorectal cancer. The proposed scGSL model demonstrated robust performance, achieving an average accuracy of 84.83%, precision of 86.23%, recall of 81.51%, and an F1 score of 80.92% across all datasets. These metrics represent a significant enhancement over existing methods, which typically exhibit lower performance metrics. Additionally, by reviewing existing literature on gene interactions within the TME, the scGSL model proves to robustly identify biologically meaningful gene interactions in an unsupervised manner, validated by significant expression differences in key gene pairs across various cancers. The source code and data used in this study can be found at https://github.com/LiYuechao1998/scGSL.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Cell CommunicationSingle-Cell AnalysisSequence Analysis, RNANeoplasmsHumansSingle-Cell Gene Expression AnalysisTumor MicroenvironmentGraph Neural Networks

Stichwörter der Autor:innen

cell type annotationcell–cell communicationligand-receptor interactionssingle-cell RNA sequencingtumor microenvironment

Autor:innen und Institutionen

  • School of Computer Science, Northwestern Polytechnical University, No.1 Dongxiang Road, Chang'an District, Xi'an 710129, Shaanxi, China.
  • College of Big Data and Internet, Shenzhen Technology University, No.3002 Lantian Road, Pingshan District, Shenzhen 518118, Guangdong, China.
  • College of Computer Science and Technology, Jilin University, No.2699 Qianjin Street, Chaoyang District, Changchun 130012, Jilin, China.
  • School of Computer Science and Technology, China University of Mining and Technology, No.1 Daxue Road, Tongshan District, Xuzhou 221116, Jiangsu, China.
  • Research Office, City University of Hong Kong (Dongguan), No.8 Gaoxiong Road, Songshan Lake High-tech Industrial Development Zone, Dongguan 523808, Guangdong, China.
  • Research and Development Institute, Northwestern Polytechnical University, No.45 Gaoxin South 9th Road, Nanshan District, Shenzhen 518063, Guangdong, China.

ORCID-Profile: Yue-Chao Li, Xin-Fei Wang, Meng-Meng Wei, Lei Wang, Zhi-An Huang, Yu-An Huang, Zhu-Hong You

Förderung

  • Guangdong Basic and Applied Basic Research Foundation
  • Fundamental Research Funds for the Central Universities
  • Neural Science Foundation of Shaanxi Province
  • National Natural Science Foundation of China
  • National Science Fund for Distinguished Young Scholars of China

Zitieren

APA

Li, Y. C., Wong, L., You, H. R., Wang, X. F., Wei, M. M., Wang, L., Huang, Z. A., Huang, Y. A., & You, Z. H. (2026). Biosemantic graph transfer learning for cross-patient cell type annotation and cell–cell interaction analysis in tumor single-cell RNA sequencing. Briefings in Bioinformatics, 27(6). https://doi.org/10.1093/bib/bbag587

Vancouver

Li YC, Wong L, You HR, Wang XF, Wei MM, Wang L, et al. Biosemantic graph transfer learning for cross-patient cell type annotation and cell–cell interaction analysis in tumor single-cell RNA sequencing. Briefings in Bioinformatics. 2026;27(6). doi:10.1093/bib/bbag587

BibTeX
@article{Li2026,
  author = {Li, Yue-Chao and Wong, Leon and You, Hai-Ru and Wang, Xin-Fei and Wei, Meng-Meng and Wang, Lei and Huang, Zhi-An and Huang, Yu-An and You, Zhu-Hong},
  title = {Biosemantic graph transfer learning for cross-patient cell type annotation and cell–cell interaction analysis in tumor single-cell RNA sequencing},
  journal = {Briefings in Bioinformatics},
  year = {2026},
  volume = {27},
  number = {6},
  doi = {10.1093/bib/bbag587}
}
Kostenlos starten →