Consensus representation of multiple cell-cell graphs from gene signaling pathways for cell type annotation

Area of Science:

• Computational Biology
• Genomics
• Bioinformatics

Background:

• Single-cell RNA sequencing (scRNA-seq) reveals tissue heterogeneity but faces challenges in accurate cell type annotation.
• Limitations include marker specificity, batch effects, and insufficient spatial/interaction data, hindering traditional annotation methods.

Purpose of the Study:

• To develop a robust computational framework for accurate cell type annotation in scRNA-seq data.
• To integrate gene expression with pathway activity for improved cell type identification.

Main Methods:

• Proposed scMCGraph, a framework integrating gene expression and pathway activity.
• Constructed multiple pathway-specific views from gene expression and pathway databases.
• Integrated views into a consensus graph for cell type annotation.

Main Results:

• scMCGraph demonstrated exceptional robustness and accuracy across cross-platform, cross-time, and cross-sample analyses.
• The consensus graph enhanced predictive performance for cell type prediction.
• Model performance improved with increased pathway information and complementary data from different databases.

Conclusions:

• scMCGraph significantly advances cell type annotation by incorporating pathway information.
• Pathway integration improves cell-cell graph learning and predictive accuracy.
• The framework shows consistent accuracy and robustness in various cross-dataset scenarios.