scLM

scLM identifies consensus co-expressed gene clusters across multiple single-cell RNA-seq (scRNA-seq) datasets to reveal functional gene modules and refine cell-state definitions.

Key Features:

• Consensus Clustering: Clusters genes co-expressed across multiple scRNA-seq datasets simultaneously to detect consensus modules.
• Batch Effect Mitigation: Minimizes batch effects while preserving biological variation when combining datasets from different studies or platforms.
• Tailored Co-clustering Algorithm: Employs a co-clustering algorithm specifically designed for single-cell datasets.
• Input Format: Accepts raw count data from single-cell RNA-seq as input.
• Benchmarking and Accuracy: Demonstrated superior performance in identifying biologically relevant gene clusters based on testing with simulated and experimental data.

Scientific Applications:

• Functional Module Discovery: Identify novel functional gene modules from scRNA-seq data.
• Cell-State Refinement: Refine cell-state definitions and improve resolution of cell identity and function.
• Cross-Dataset Comparative Analysis: Enable comparative analysis across studies and platforms by detecting consensus co-expression patterns.
• Cancer Research and Mechanism Discovery: Support mechanism discovery in diseases such as cancers by revealing conserved co-expression modules.

Methodology:

scLM takes raw count data as input and applies a tailored co-clustering algorithm specifically designed for single-cell datasets.