A new gene regulatory network model based on BP algorithm for interrogating differentially expressed genes of Sea...

Area of Science:

• Computational Biology
• Systems Biology
• Bioinformatics

Background:

• Traditional gene regulatory network models struggle with high-dimensional gene expression data due to complexity.
• Existing models often lack effective methods for analyzing dynamic and spatial properties of gene interactions.

Purpose of the Study:

• To develop a new computational model for constructing gene regulatory networks.
• To leverage back propagation (BP) neural networks for improved analysis of gene expression data.
• To identify differentially expressed genes and analyze network properties.

Main Methods:

• Constructed gene regulatory networks using a back propagation (BP) neural network model.
• Integrated predictive functions and network topology for model development.
• Analyzed network characteristics including average path length, clustering coefficient, degree, modularity, and density.
• Utilized Sea Urchin mRNA microarray data under varying temperatures for simulation and analysis.

Main Results:

• The BP neural network successfully simulated gene regulatory networks as complex artificial networks.
• Statistical analysis of network parameters revealed insights into gene interactions under different temperatures.
• Identified differentially expressed Sea Urchin genes associated with temperature changes by comparing network degrees.

Conclusions:

• The developed BP neural network model is effective for simulating gene regulatory networks.
• The model demonstrates capability in accurately determining differentially expressed genes.