A Consensus Gene Regulatory Network for Neurodegenerative Diseases Using Single-Cell RNA-Seq Data.

Area of Science:

• Systems Biology
• Computational Biology
• Neuroscience

Background:

• Gene regulatory network (GRN) inference is crucial for understanding complex diseases like neurodegeneration.
• Existing GRN inference methods have limitations due to their underlying assumptions, restricting biological insights.
• A consensus approach is needed to overcome individual method limitations and enhance GRN analysis robustness.

Purpose of the Study:

• To develop and evaluate an ensemble GRN method by integrating multiple state-of-the-art algorithms.
• To address the limitations of individual GRN inference methods for improved biological understanding.
• To identify potential key regulators and subnetworks in neurodegeneration using a consensus GRN.

Main Methods:

• Selected four distinct GRN inference algorithms, incorporating both static and dynamic approaches.
• Constructed a consensus GRN by identifying common gene interactions predicted by the selected algorithms.
• Applied the ensemble method to a single-cell RNA sequencing (scRNA-seq) dataset from a mouse model of neurodegeneration (CK-p25).

Main Results:

• The consensus GRN revealed significant overlap in gene interactions across different algorithms for the studied dataset.
• Identified potential key regulators and critical subnetworks within the neurodegenerative context.
• Demonstrated the necessity and effectiveness of an ensemble approach for robust GRN inference.

Conclusions:

• An ensemble GRN approach combining diverse algorithms provides more robust and biologically relevant insights than single methods.
• This study validates the creation of a consensus network for analyzing complex biological systems like neurodegeneration.
• The findings highlight the potential of integrated GRN methods for discovering critical regulators in disease.