A Swarm Intelligence Framework for Reconstructing Gene Networks: Searching for Biologically Plausible Architectures

Abstract

In this paper, we investigate the problem of reverse engineering the topology of gene regulatory networks from temporal gene expression data. We adopt a computational intelligence approach comprising swarm intelligence techniques, namely particle swarm optimization (PSO) and ant colony optimization (ACO). In addition, the recurrent neural network (RNN) formalism is employed for modeling the dynamical behavior of gene regulatory systems. More specifically, ACO is used for searching the discrete space of network architectures and PSO for searching the corresponding continuous space of RNN model parameters. We propose a novel solution construction process in the context of ACO for generating biologically plausible candidate architectures. The objective is to concentrate the search effort into areas of the structure space that contain architectures which are feasible in terms of their topological resemblance to real-world networks. The proposed framework is initially applied to the reconstruction of a small artificial network that has previously been studied in the context of gene network reverse engineering. Subsequently, we consider an artificial data set with added noise for reconstructing a subnetwork of the genetic interaction network of S. cerevisiae (yeast). Finally, the framework is applied to a real-world data set for reverse engineering the SOS response system of the bacterium Escherichia coli. Results demonstrate the relative advantage of utilizing problem-specific knowledge regarding biologically plausible structural properties of gene networks over conducting a problem-agnostic search in the vast space of network architectures.

Analysis of Microarray Data using Artificial Intelligence Based Techniques

“…These techniques can be clubbed into different groups, such as Boolean networks (Liang et al, 1998;Akutsu et al, 1999;Shmulevich et al, 2002;Martin et al, 2007;Raza and Jaiswal, 2013;Raza and Parveen, 2013), Bayesian networks (Friedman et al, 2000;Husmeier, 2003), Petri nets (Koch et al, 2005;Remy et al, 2006), linear and non-linear ordinary differential equations (ODEs) (Chen et al, 1999;Tyson et al, 2002;De Jong and Page, 2008), machine learning approaches (Weaver et al, 1999;Kim et al, 2000;Vohradský, 2001;Keedwell et al, 2002;Huang et al, 2003;Tian and Burrage, 2003;Zhou et al, 2004;Xu et al, 2004;Hu et al, 2006;Jung and Cho, 2007;Xu et al, 2007a,b;Chiang and Chao, 2007;Lee and Yang, 2008;Datta et al, 2009;Zhang et al, 2009;Maraziotis et al, 2010;Ghazikhani et al, 2011;Liu et al, 2011;Kentzoglanakis and Poole, 2012;Noman et al, 2013), etc.…”

When Neural Network Computation Meets Evolutionary Computation: A Survey

“…In addition to the parameters mentioned above, EC algorithms are available to adapt other parameters of NNs such as, timescale [16,18], parameter of transfer function [19], smoothing factor [31]. Moreover, EC algorithms can also optimize a method that is applied to improve an NN.…”

“…Apart from the fields mentioned above, EC-NNs are used in other fields as well, where they can be applied to forecast energy consumption of cloud computing [15], reconstruct the topology of gene regulatory network [16], design custom-made fractal antennas [33], optimize large scale problem [20] as well as employed in Spatio-Temporal system identification [32], and an online modeling algorithm [25].…”

“…The paper [16] proposed a novel hybrid ant colony optimization (ACO) and PSO method to reconstruct the topology of gene regulatory networks. The paper [13] proposed CCPSO which combine culture algorithm for accelerating the search and enhancing global search capacity.…”

Imputation of Missing Gene Expressions for DNA Microarray Using Particle Swarm Optimization

“…Data generated by the high throughput technology of Microarray is useful in many applications like identification of most significant genes, clustering [1], recognition of significant patterns in gene expressions, modeling of Gene Regulatory Network (GRN) [2], drug designing, classification [3], etc. But microarray creates predominant problem of generating missing values in gene expression data set.…”