A new and effective two-step clustering approach for single cell RNA sequencing data

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Single-cell RNA sequencing (scRNA-seq) generates vast datasets crucial for understanding biological complexity.
• Analyzing scRNA-seq data, particularly cell clustering, is vital for fields like disease pathogenesis and drug resistance research.
• Current cell clustering methods require improvement for enhanced performance.

Purpose of the Study:

• To develop a novel, effective two-step clustering approach for scRNA-seq data analysis.
• To improve the accuracy and performance of cell clustering in scRNA-seq studies.

Main Methods:

• Introduced Two-Step Clustering (TSC), a new scRNA-seq data analysis method.
• TSC employs a two-step strategy: hierarchical clustering of core cells followed by assignment of non-core cells.
• Core cells are identified as those near cluster centers, while non-core cells are in boundary areas.

Main Results:

• TSC demonstrated superior performance compared to state-of-the-art methods across 12 real scRNA-seq datasets.
• The two-step clustering strategy effectively handles cell heterogeneity in scRNA-seq data.
• Experiments confirmed TSC's ability to accurately infer cell clusters.

Conclusions:

• TSC is a powerful and effective tool for scRNA-seq data analysis.
• The proposed two-step clustering strategy enhances the reliability of cell clustering.
• TSC offers a valuable advancement for researchers utilizing scRNA-seq technology.