scBatch: batch-effect correction of RNA-seq data through sample distance matrix adjustment.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Batch effects are a common challenge in high-throughput sequencing data analysis.
• Existing methods often fail to adequately correct batch effects, particularly in single-cell RNA sequencing (RNA-seq).
• These uncorrected effects can lead to erroneous conclusions in downstream analyses.

Purpose of the Study:

• To introduce scBatch, a novel numerical algorithm for robust batch-effect correction.
• To enhance both clustering and gene differential expression analysis in bulk and single-cell RNA-seq data.
• To provide a method that is not limited by assumptions regarding batch-effect generation mechanisms.

Main Methods:

• Development of a numerical algorithm, scBatch, for batch-effect correction.
• Application of scBatch to both bulk and single-cell RNA sequencing datasets.
• Evaluation of scBatch performance using simulations and real-world data analyses.

Main Results:

• scBatch demonstrates superior performance compared to benchmark batch-effect correction methods.
• The algorithm effectively improves clustering and gene differential expression analysis.
• Simulations and real data analyses confirm the efficacy of scBatch.

Conclusions:

• scBatch offers a powerful and versatile solution for batch-effect correction in RNA sequencing data.
• The method enhances the reliability of biological insights derived from sequencing experiments.
• scBatch is available as an R package for widespread use in the research community.