Identification of Human Cell Cycle Phase Markers Based on Single-Cell RNA-Seq Data by Using Machine Learning Methods.

Area of Science:

• Molecular Biology
• Genomics
• Bioinformatics

Background:

• The cell cycle, comprising G1, S, G2, and M phases, is crucial for cell growth and division.
• Accurate cell cycle phase determination is vital for cancer research and targeted therapies.
• Current methods are complex, time-consuming, and lack scalability.

Purpose of the Study:

• To develop an efficient and scalable method for cell cycle phase determination using single-cell RNA sequencing data.
• To identify novel gene biomarkers and classification rules for distinguishing cell cycle phases.
• To overcome limitations of existing cell cycle detection techniques.

Main Methods:

• Utilized single-cell RNA sequencing (scRNA-seq) data.
• Employed Boruta and feature ranking algorithms (mRMR, MCFS, SHAP by LightGBM) for biomarker identification.
• Constructed efficient classifiers using advanced machine learning algorithms.

Main Results:

• Identified essential gene biomarkers for cell cycle phase classification.
• Developed a series of classification rules to distinguish between cell cycle phases.
• Demonstrated the effectiveness of the novel method for large-scale cell cycle analysis.

Conclusions:

• The study presents a novel, efficient, and scalable method for cell cycle phase determination.
• New potential cell cycle-related genes and biomarkers were identified.
• This contributes to a better understanding of cell cycle regulation and its implications in disease.