Inferring the regulation dynamics of oscillatory networks from scRNA-seq data.

Area of Science:

• Genomics
• Systems Biology
• Developmental Biology

Background:

• Cell cycle processes are fundamental to cell fate determination and disease.
• Current gene regulatory network (GRN) inference methods often overlook the cyclical nature of biological processes.
• Accurate GRN inference is essential for understanding complex biological regulation.

Purpose of the Study:

• To enhance GRN inference accuracy by accounting for the oscillatory nature of the cell cycle.
• To test the hypothesis that constraining cell cycle positions improves GRN inference.
• To investigate cell cycle regulation in mouse retinal progenitor cells.

Main Methods:

• Evaluation of eight representative GRN inference methods.
• Application of three selected methods to a mouse retinal progenitor single-cell gene expression dataset.
• Integration of cell cycle positions inferred by Tricycle for improved temporal ordering.

Main Results:

• Incorporating cell cycle positions significantly improved GRN inference accuracy compared to using experimental times.
• Improvements were particularly notable for early progenitor cells, suggesting intrinsic cell cycle regulation.
• The study demonstrated the utility of Tricycle for inferring cell cycle states.

Conclusions:

• Constraining GRN inference with cell cycle positions enhances accuracy, especially for oscillatory processes.
• This approach offers a promising framework for advancing the understanding of gene regulation in cyclic biological systems.
• Future research should explore integrating oscillatory dynamics into causal inference for broader biological insights.