SSCC: A Novel Computational Framework for Rapid and Accurate Clustering Large-scale Single Cell RNA-seq Data.

Area of Science:

• Computational biology
• Genomics
• Bioinformatics

Background:

• Single-cell RNA sequencing (scRNA-seq) generates massive datasets, posing computational challenges for traditional clustering methods.
• Accurate and efficient clustering algorithms are crucial for extracting meaningful biological insights from scRNA-seq data.
• Existing methods struggle with the scalability and accuracy required for large-scale scRNA-seq analyses.

Purpose of the Study:

• To develop a novel clustering framework for large-scale scRNA-seq data analysis.
• To improve both the accuracy and computational efficiency of scRNA-seq data clustering.
• To provide a robust and scalable solution for analyzing high-volume single-cell data.

Main Methods:

• Proposed Spearman subsampling-clustering-classification (SSCC) framework.
• Utilized random projection and feature construction techniques.
• Benchmarked SSCC against state-of-the-art algorithms on multiple real-world scRNA-seq datasets.

Main Results:

• SSCC significantly enhances clustering accuracy, robustness, and computational efficacy.
• On a dataset of 68,578 human blood cells, SSCC improved accuracy by 20% and achieved 50-fold acceleration compared to SC3.
• SSCC demonstrated up to 3-fold accuracy improvement over k-means while consuming less memory.

Conclusions:

• SSCC provides a computationally efficient and accurate framework for clustering large-scale scRNA-seq data.
• The proposed method addresses the pressing need for scalable analytical tools in single-cell genomics.
• An R implementation of SSCC is publicly available for broader research use.