Deciphering Cell Cycle Dynamics and Cell States in Single-cell RNA-seq data with SPAE

Area of Science:

• Genomics
• Computational Biology
• Bioinformatics

Background:

• Single-cell RNA sequencing (scRNA-seq) reveals cellular heterogeneity and complex biological processes.
• Analyzing cell cycle dynamics in scRNA-seq data is challenging due to data complexity and subtle cell state differences.

Purpose of the Study:

• To develop an accurate and robust method for characterizing cell cycle dynamics and cell states in scRNA-seq data.
• To address the limitations of existing methods in cell cycle analysis.

Main Methods:

• Development of the integrated Sinusoidal and Piecewise AutoEncoder (SPAE), an autoencoder-based piecewise linear model.
• Application of SPAE to scRNA-seq data for cell cycle characterization and effect removal.

Main Results:

• SPAE demonstrated improved accuracy and robustness in characterizing cell cycle dynamics compared to existing methods.
• SPAE accurately predicted cancer cell cycle transitions.
• SPAE effectively facilitated the removal of cell cycle effects from gene expression data.

Conclusions:

• SPAE is a powerful tool for analyzing cell cycle dynamics in scRNA-seq data.
• The method enhances the understanding of cellular heterogeneity and disease pathogenesis.
• SPAE is available for non-commercial use.