Batch Effect Correction

In our sn/scRNA-Seq methods tumor toolbox paper, we integrated patient datasets using the Canonical Correlation Analysis and dynamic time warping algorithm implemented in Seurat v2 in order to identify shared cell types. An example of integrating different datasets is provided in the Seurat tutorial.

There are many other available batch effect correction/data integration algorithms and three others we commonly use include: LIGER (example with COVID-19 lung tissue), Harmony (example with COVID-19 lung tissue), and CONOS.