sc-batch-integration

To remove batch effects in single-cell RNA-seq data, use batch integration methods like Harmony or scVI. These algorithms eliminate technical variation across multi-sample datasets while preserving true biological differences for unified downstream analysis.

The best way to integrate multi-sample scRNA-seq datasets is using algorithms like Seurat CCA/RPCA, BBKNN, and fastMNN. These methods harmonize data from different experimental conditions to enable unified downstream analysis like identifying cell types.

Yes, scVI batch integration works with scanpy and anndata. Running this requires Python 3.11+ and utilizes libraries including scanpy, anndata, and scvi-tools to process and harmonize single-cell RNA sequencing data effectively.

You need data harmonization for scRNA-seq experiments when combining datasets run on different days or with different reagent lots. This process removes the technical variation that obscures true biological differences across conditions.

This Skill supports both Harmony and fastMNN for single-cell data integration. You can apply these methods independently to integrate multiple scRNA-seq datasets and remove technical variation while preserving biological signals.