Inferring gene regulatory networks by singular value decomposition and gravitation field algorithm

Area of Science:

• Systems Biology
• Computational Biology
• Bioinformatics

Background:

• Reconstructing gene regulatory networks (GRNs) is crucial in systems biology but computationally challenging.
• Existing GRN inference algorithms have limitations in effectiveness and efficiency.
• Gene expression data is a primary source for inferring GRN structures.

Purpose of the Study:

• To develop a novel and effective algorithm for inferring GRNs from gene expression data.
• To improve the accuracy and efficiency of GRN reconstruction.
• To address the limitations of current GRN inference methods.

Main Methods:

• A novel inference algorithm based on a differential equation model was proposed.
• Singular value decomposition (SVD) was used for data decomposition and defining the solution space.
• A modified gravitation field algorithm (GFA) was employed to optimize the differential equation model and search for the best network structure.
• The algorithm was validated using simulated scale-free networks and a real benchmark GRN dataset.

Main Results:

• The proposed algorithm demonstrated superior performance in reconstructing GRNs compared to Bayesian methods and traditional differential equation models.
• Cross-validation confirmed the high effectiveness and accuracy of the novel algorithm.
• The modified GFA effectively optimized the differential equation model criteria for GRN inference.

Conclusions:

• The developed algorithm offers a significant advancement in GRN reconstruction from gene expression data.
• The combination of SVD and modified GFA provides a robust framework for accurate and efficient GRN inference.
• This method outperforms existing approaches, offering a valuable tool for systems biology research.