batchelor

batchelor implements methods for batch correction of single-cell RNA sequencing data, including mutually nearest neighbor detection, efficient variants of linear regression on log-expression values, global rescaling to remove depth differences between batches, and principal components analysis robust to differences in cell numbers across batches.

Key Features:

• Mutual nearest neighbor batch correction: Detects mutually nearest neighbors to correct batch effects in single-cell RNA sequencing data.
• Linear regression batch correction: Provides several efficient variants of linear regression of log-expression values for batch correction.
• Global rescaling: Removes differences in sequencing depth between batches.
• Robust principal components analysis: Performs PCA that is robust to differences in the numbers of cells across batches.

Scientific Applications:

• Single-cell RNA sequencing batch correction: Corrects technical batch effects in single-cell RNA sequencing datasets to enable integrative analysis.
• Cross-batch data integration: Enables combination of single-cell datasets from different experimental batches while preserving biological variation.

Methodology:

Methods based on detecting mutually nearest neighbors; efficient variants of linear regression of log-expression values; global rescaling; and principal components analysis robust to differences in cell numbers across batches.