scMerge

scMerge integrates multiple single-cell RNA-seq datasets using factor analysis of stably expressed genes and pseudoreplicates to remove unwanted variations and enable accurate biological interpretation.

Key Features:

• Stably Expressed Gene Identification: Identifies stably expressed genes (SEGs) across datasets to serve as anchors for batch effect correction.
• Pseudoreplicate Detection: Implements replication-identification methods to recognize pseudoreplicates across multiple single-cell datasets.
• Factor Analysis-Based Integration: Leverages factor analysis to remove unwanted variations and merge multiple single-cell RNA-seq datasets.
• Complete Pipeline Implementation: Provides all necessary functions for SEG identification, replication identification, and merging of single-cell data.

Scientific Applications:

• Batch Effect Removal: Removes batch effects and other unwanted variations that hinder accurate biological interpretation of single-cell RNA-seq data.
• Cell Type Separation: Improves cell type separation across integrated datasets compared to published methods.
• Development Trajectory Inference: Enhances biological discovery through robust data integration, enabling inference of developmental trajectories in complex tissues such as liver.
• Multi-Dataset Biological Discovery: Supports concerted examination of multiple single-cell RNA-seq collections to uncover biological insights not accessible from individual datasets.

Methodology:

scMerge uses factor analysis of stably expressed genes (SEGs) and pseudoreplicates across datasets to remove unwanted factors and merge multiple single-cell RNA-seq data.