STACAS

STACAS improves integration of single-cell RNA sequencing (scRNA-seq) datasets in Seurat by identifying subtype-specific integration anchors, filtering aberrant anchors with a quantitative distance measure, constructing guide trees, and correcting batch effects while preserving biological variability.

Key Features:

• Anchor Identification: STACAS identifies integration anchors between scRNA-seq datasets, prioritizing anchors shared by subsets of cell types.
• Batch Effect Correction: It corrects batch effects across datasets while preserving essential biological variability.
• Aberrant Anchor Filtering: STACAS employs a quantitative distance measure to detect and filter aberrant integration anchors.
• Guide Tree Construction: The method constructs guide trees to define optimal hierarchical integration orders for partially overlapping cell populations.

Scientific Applications:

• Single-cell atlas construction: Integrating diverse scRNA-seq datasets to build comprehensive single-cell atlases.
• Integration of public scRNA-seq datasets: Merging heterogeneous publicly available scRNA-seq datasets that exhibit partial overlap in cell types or states.
• Comparative analysis of partially overlapping populations: Enabling comparative studies across datasets that share only subsets of cell types or states.

Methodology:

Within the Seurat framework, STACAS identifies subtype-specific integration anchors, applies a quantitative distance measure to filter aberrant anchors, constructs guide trees for integration, and performs batch effect correction to preserve biological variability.