SCMcluster

SCMcluster performs high-precision clustering of single-cell RNA sequencing (scRNA-seq) data by integrating CellMarker and PanglaoDB marker gene sets with expression profiles to improve feature extraction and cell-type identification.

Key Features:

• Integration of Marker Gene Databases: Incorporates CellMarker and PanglaoDB to enhance feature extraction from scRNA-seq data using curated cell marker sets.
• Ensemble Clustering Model: Constructs an ensemble clustering model from a consensus matrix that aggregates results from multiple clustering methods to improve robustness and precision.
• Performance Evaluation: Benchmarked against eight popular clustering algorithms on two scRNA-seq datasets from human and mouse tissues, demonstrating superior feature extraction and clustering accuracy.

Scientific Applications:

• Single-cell genomics: Enables precise cell-type identification and characterization in scRNA-seq studies.
• Developmental biology: Resolves cellular differentiation states and composition during development.
• Cancer research: Identifies tumor subpopulations and intratumoral heterogeneity from scRNA-seq data.
• Immunology: Characterizes immune cell types and activation states in complex tissues.
• Cross-species tissue analysis: Applied to human and mouse scRNA-seq datasets to compare and validate cell populations across species.

Methodology:

Combine scRNA-seq data with curated cell marker databases for feature extraction and construct an ensemble clustering model using a consensus matrix that aggregates multiple clustering approaches.