Application of Meta-Heuristics on Reconstructing Gene Regulatory Network: A Bayesian Model Approach

Suman Mitra,Surama Biswas,Sriyankar Acharyya
DOI: https://doi.org/10.1080/03772063.2021.1946433
IF: 1.8768
2021-07-04
IETE Journal of Research
Abstract:Reconstruction of the Gene Regulatory Network (GRN) is important in understanding the functionalities of a living cell. The reconstruction of GRN is based on microarray gene expression data. In this paper, Bayesian Network (BN) model is used for GRN reconstruction. Here, to train BN-modelled GRN from gene expression data, four meta-heuristic techniques, namely, Simulated Annealing, Tabu Simulated Annealing, Binary Particle Swarm Optimisation (BPSO), and Binary Differential Evolution (BDE) have been applied. The algorithms have been executed on two synthetic datasets and two real datasets. The comparison of their performances demonstrates that BDE- and BPSO-based algorithms are superior to other algorithms. These two efficient algorithms have outperformed the state-of-the-art methods in consideration of F1 score.
telecommunications,engineering, electrical & electronic
What problem does this paper attempt to address?