Introduction

  • What is Batch Effect?

Batch effect occurs when differences among sets of samples are due to technical arrangements rather than biological factors, resulting in inaccurate conclusions. Single-cell data often comes from different experiments with variations in timing, personnel, reagents, equipment, and technology. This leads to batch effects, which batch effect correction aims to fix when combining cells from various batches or studies.

  • How to detect Batch Effect?

In the realm of scRNAseq analysis, a suite of tools aids in detecting batch effects, a crucial step to ensure data accuracy. Among these, Harmony stands out for its automatic batch effect identification and correction capabilities Chen et al., (2021). By seamlessly integrating datasets from different batches, Harmony reduces technical noise and improves clustering accuracy, all without requiring complex parameter tuning .

Dataset

The data used in this report was obtained from the 10X dataset, which comprises three specific dataset tables listed below:

These datasets included peripheral blood mononuclear cells (PBMCs) sourced from a healthy donor. These datasets were generated using varying chemistry platforms and encompassed different cell quantities.

Batch effect correction

Before Using Harmony

The UMAP visualization displays the cellular clusters based on individual datasets before to the process of integration.

After Using Harmony

The UMAP visualization displays the cellular clusters based on individual datasets after to the process of integration.

Annotation

During the Cell Type Annotation stage, SingleR was employed to carry out this procedure.

Before using Harmony

Cluster

The UMAP visualization displays the cellular clusters based on seurat clusters.

Annotation

The UMAP visualization displays the cellular clusters based on cell type annotation.

After using Harmony

Cluster

The UMAP visualization displays the cellular clusters based on seurat cluster after the process of integration.

Annotation

The UMAP visualization displays the cellular clusters based on cell type annotation

Software Catalog

Analysis Software Version
Pre - Processing data CellRanger 7.1.0
Processing Data Seurat 4.9.9.9045
Batch Effect Correction Harmony 0.1.1
Cell Type Annotation SingleR 1.8.1