Simultaneous structure discovery and parameter estimation in gene networks using a multi-objective GP-PSO hybrid...

Area of Science:

• Systems Biology
• Computational Biology
• Bioinformatics

Background:

• Understanding gene regulatory networks (GRNs) is crucial for deciphering complex biological processes.
• Automated methods are needed to infer GRN structure from experimental data.
• Plant flowering time is a key developmental trait influenced by intricate genetic interactions.

Purpose of the Study:

• To develop and validate a hybrid algorithm for automated gene network structure recovery.
• To utilize gene expression time series and phenotypic data for GRN inference.
• To approximate plant gene regulatory networks controlling flowering time.

Main Methods:

• A hybrid algorithm integrating Genetic Programming (GP) and Particle Swarm Optimisation (PSO).
• Input data includes gene expression time series and phenotypic data (plant flowering time).
• The algorithm aims to discover parsimonious GRN structures.

Main Results:

• The proposed GP-PSO hybrid algorithm successfully recovered gene network structures.
• Simulated gene expression and flowering times closely matched input data.
• Demonstrated efficacy in modeling flowering time control in Arabidopsis thaliana.

Conclusions:

• The hybrid GP-PSO approach offers an effective method for automated GRN inference.
• This computational strategy aids in understanding genetic control of plant development.
• The findings contribute to advancing systems biology and synthetic biology applications.