Reconstruction of Gene Regulatory Network Using Recurrent Neural Network Model: A Harmony Search Approach
Gene regulatory network (GRN) is an artificial network comprising a gene set and their mutual interactions. One gene interacts with other through protein production which is called ‘gene expression.’ The computational reconstruction of GRN is made using gene expression data where gene expression values are provided in a matrix. Here, a recurrent neural network (RNN) model is used to represent the GRN. In this paper, a meta-heuristic algorithm, namely harmony search (HS), has been modified and applied to train the network parameters in reconstructing GRN. The proposed HS-GRN algorithm is applied to a standard artificial gene expression dataset, and the experimental results are compared with several other methods. It shows that HS-GRN performs better than the literature methods on this dataset. Furthermore, HS-GRN has also been applied to two real gene datasets, namely IRMA-ON and IRMA-OFF. The experimental results state that HS-GRN performs satisfactorily well in real datasets also.
KeywordsGene regulatory network Recurrent neural network Gene expression Harmony search Meta-heuristics
We are thankful to TEQIP-III Maulana Abul Kalam Azad University of Technology (MAKAUT), West Bengal, India, for supporting our research.
- 1.Xu, R., Wunsch II, D., Frank, R.L.: Inference of genetic regulatory networks with recurrent neural network models using particle swarm optimization. IEEE/ACM Trans. Comput. Biol. Bioinform. 4(4), 681–692 (2007)Google Scholar
- 5.Biswas, S., Acharyya, S.: A Bi-objective RNN model to reconstruct gene regulatory network: a modified multi-objective simulated annealing approach. IEEE/ACM Trans. Comput. Biol. Bioinform. (99), 1–8 (2017)Google Scholar
- 6.Donoho, D.L.: High-dimensional data analysis: the curses and blessings of dimensionality. AMS Math. Chall. Lect. 1, 1–32 (2000)Google Scholar
- 9.Kordmahalleh, M.M., Sefidmazgi, M.G., Harrison, S.H., Homaifar, A.: Identifying time-delayed gene regulatory networks via an evolvable hierarchical recurrent neural network. BioData Min. 10(1), 29 (2017)Google Scholar
- 10.Morshed, N., Chetty, M.: Reconstructing genetic networks with concurrent representation of instantaneous and time-delayed interactions. In: Proceedings of the 2011 IEEE Congress on Evolutionary Computation (CEC), pp. 1840–1847 (2011)Google Scholar
- 11.Morshed, N., Chett, M.: Information theoretic dynamic Bayesian network approach for reconstructing genetic networks. In: Proceedings of the 11th IASTED International Conference on Artificial Intelligence and Applications, pp. 236–43 (2011)Google Scholar
- 13.Gatta, G.D., Bansalm, M., Ambesi-Impiombato, A., Antonini, D., Missero, C., di Bernardo, D.: Direct targets of the TRP63 transcription factor revealed by a combination of gene expression profiling and reverse engineering. Genome Res. 18(6), 939–948 (2008)Google Scholar
- 14.Margolin, A., Nemenman, I., Basso, K., Wiggins, C., Stolovitzky, G., Favera, R.D.: ARACNE: an algorithm for the reconstruction of gene regulatory networks in a mammalian cellular context. BMC Bioinform. 7(1), S7 (2006)Google Scholar