Gene regulatory networks using bat algorithm inspired particle swarm optimization
01 Dec 2015-pp 387-390
TL;DR: A statistical framework based on a novel bat algorithm inspired particle swarm optimisation algorithm for the reconstruction of gene regulatory networks from temporal gene expression data and results obtained suggest that the proposed methodology can infer the underlying network structures with a better degree of success.
Abstract: Here, we have proposed a statistical framework based on a novel bat algorithm inspired particle swarm optimisation algorithm for the reconstruction of gene regulatory networks from temporal gene expression data. The recurrent neural network formalism has been implemented to extract the underlying dynamics from time series microarray datasets accurately. The proposed swarm intelligence framework has been used for optimising the parameters of the recurrent neural network model. Preliminary research with the proposed methodology has been done on a small, artificial network and the experimental (in vivo) microarray data of the SOS DNA repair network of Escherichia coli. Results obtained suggest that the proposed methodology can infer the underlying network structures with a better degree of success.
Citations
More filters
26 Apr 2017
TL;DR: A novel quantum computing based technique for the reverse engineering of gene regulatory networks from time-series genetic expression datasets is proposed, suggesting that quantum computing technique significantly reduces the computational time, retaining the accuracy of the inferred gene Regulatory networks to a comparatively satisfactory level.
Abstract: The accurate reconstruction of gene regulatory networks from temporal gene expression data is crucial for the identification of genetic inter-regulations at the cellular level. This will help us to comprehend the working of living entities properly. Here, we have proposed a novel quantum computing based technique for the reverse engineering of gene regulatory networks from time-series genetic expression datasets. The dynamics of the temporal expression profiles have been modelled using the recurrent neural network formalism. The corresponding training of model parameters has been realised with the help of the proposed quantum computing methodology based concepts. This is based on entanglement and decoherence concepts. The application of quantum computing technique in this domain of research is comparatively new. The results obtained using this technique is highly satisfactory. We have applied it to a 4-gene artificial genetic network model, which was previously studied by other researchers. Also, a 10-gene and a 20-gene genetic network have been studied using the proposed technique. The obtained results suggest that quantum computing technique significantly reduces the computational time, retaining the accuracy of the inferred gene regulatory networks to a comparatively satisfactory level.
2 citations
References
More filters
04 Oct 1995
TL;DR: The optimization of nonlinear functions using particle swarm methodology is described and implementations of two paradigms are discussed and compared, including a recently developed locally oriented paradigm.
Abstract: The optimization of nonlinear functions using particle swarm methodology is described. Implementations of two paradigms are discussed and compared, including a recently developed locally oriented paradigm. Benchmark testing of both paradigms is described, and applications, including neural network training and robot task learning, are proposed. Relationships between particle swarm optimization and both artificial life and evolutionary computation are reviewed.
14,477 citations
"Gene regulatory networks using bat ..." refers methods in this paper
...Here, in BA-PSO, the initial velocity of the particles has been chosen as 0 similar to that done in BA....
[...]
...The training of the RNN model has been implemented using our proposed BA-PSO algorithm....
[...]
...To reduce the computational burden, these can safely be decomposed (without any loss of generality) into N optimisation subproblems: where (N+2) parameters are to be estimated for each of the N genes individually by PSO....
[...]
...The other optimises the RNN model parameters using the proposed BA-PSO algorithm such that the trained model can mimic the original network dynamics accurately....
[...]
...Also, PSO inspired by the comparatively much newer BA has been used to train the RNN model parameters such that the predicted networks could reproduce the dynamics of the given datasets as closely as possible....
[...]
TL;DR: The hypothesis that contemporary organisms are also randomly constructed molecular automata is examined by modeling the gene as a binary (on-off) device and studying the behavior of large, randomly constructed nets of these binary “genes”.
4,250 citations
"Gene regulatory networks using bat ..." refers methods in this paper
...Several different approaches such as Boolean networks [5], and S-systems [6] are used to construct GRNs....
[...]
Posted Content•
TL;DR: The Bat Algorithm as mentioned in this paper is based on the echolocation behavior of bats and combines the advantages of existing algorithms into the new bat algorithm to solve many tough optimization problems.
Abstract: Metaheuristic algorithms such as particle swarm optimization, firefly algorithm and harmony search are now becoming powerful methods for solving many tough optimization problems. In this paper, we propose a new metaheuristic method, the Bat Algorithm, based on the echolocation behaviour of bats. We also intend to combine the advantages of existing algorithms into the new bat algorithm. After a detailed formulation and explanation of its implementation, we will then compare the proposed algorithm with other existing algorithms, including genetic algorithms and particle swarm optimization. Simulations show that the proposed algorithm seems much superior to other algorithms, and further studies are also discussed.
3,528 citations
23 Apr 2010
TL;DR: The Bat Algorithm as mentioned in this paper is based on the echolocation behavior of bats and combines the advantages of existing algorithms into the new bat algorithm to solve many tough optimization problems.
Abstract: Metaheuristic algorithms such as particle swarm optimization, firefly algorithm and harmony search are now becoming powerful methods for solving many tough optimization problems. In this paper, we propose a new metaheuristic method, the Bat Algorithm, based on the echolocation behaviour of bats. We also intend to combine the advantages of existing algorithms into the new bat algorithm. After a detailed formulation and explanation of its implementation, we will then compare the proposed algorithm with other existing algorithms, including genetic algorithms and particle swarm optimization. Simulations show that the proposed algorithm seems much superior to other algorithms, and further studies are also discussed.
3,162 citations
Book•
01 Sep 2000
TL;DR: This work presents a graphical representation of biochemical systems, a sequence of models describing purine metabolism, and a model of the tricarboxylic acid cycle in Dictyostelium discoideum, which shows the importance of knowing the initial steps of the Glycolytic-Glycogenolytic pathway.
Abstract: Preface Introduction 1 Graphical representation of biochemical systems 2 Models of biochemical systems 3 From maps to equations 4 Computer simulation 5 Parameter estimation 6 Analytical steady-state evaluation 7 Sensitivity analysis 8 Case study 1 - Anaerobic fermentation pathway in Saccharomyces cerevisiae 9 Case study 2 - diagnosis and refinement of a model of the tricarboxylic acid cycle in Dictyostelium discoideum 10 Case study 3 - A sequence of models describing purine metabolism 11 Case study 4 - Algebraic analysis of the initial steps of the Glycolytic-Glycogenolytic pathway in perfused rat liver 12 Epilogue-Canonical modeling beyond biochemistry Appendix Hints and solutions References Author index Subject index
617 citations
"Gene regulatory networks using bat ..." refers methods in this paper
...Several different approaches such as Boolean networks [5], and S-systems [6] are used to construct GRNs....
[...]