Single-cell data clustering based on sparse optimization and low-rank matrix factorization.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Unsupervised clustering is crucial for single-cell RNA-sequencing (scRNA-seq) data analysis.
• Accurate cell cluster prediction remains a significant challenge in scRNA-seq studies.
• Existing methods face limitations in precisely classifying diverse cell populations.

Purpose of the Study:

• To introduce a novel algorithm for scRNA-seq data clustering.
• To enhance the accuracy of cell type identification in scRNA-seq datasets.
• To provide a robust computational tool for scRNA-seq data analysis.

Main Methods:

• Development of a new algorithm named scSO.
• Application of Sparse Optimization and low-rank matrix factorization.
• Analysis of multiple benchmark scRNA-seq datasets.

Main Results:

• scSO accurately predicted the number of cell clusters, closely matching reference cell types.
• The algorithm demonstrated high accuracy in classifying individual cells.
• Performance was validated across diverse benchmark datasets.

Conclusions:

• scSO offers a powerful approach for cell clustering in scRNA-seq data.
• The method effectively aids researchers in distinguishing cell types.
• scSO provides a valuable tool for advancing scRNA-seq data interpretation.