A Parallel Approach for Accelerated Parameter Identification of Gene Regulatory Networks Tariq Saeed and Jamil Ahmad Research Centre for Modeling and Simulation (RCMS), National University of Sciences and Technology (NUST), Pakistan {tariq,jamil.ahmad}@rcms.nust.edu.pk http://www.rcms.nust.edu.pk Abstract. Model Checking is one of the formal verification methods which can be used to infer parameters of a Gene Regulatory Network (GRN) using discrete formalism of Ren´ e Thomas. However, the sequen- tial approach for identification of these logical parameters is computa- tionally intensive and takes lot of processing time, depending on number of genes and range of their expression levels in a network. In this paper, we present an efficient approach for this problem, based on parallel com- puting. We partition the parameter space into subsets and assign them to processing elements on a distributed memory parallel computer. The presented approach is implemented by using OpenMPI and existing tool SMBioNet. The experimental results indicate that the approach is scal- able and achieves 7X speed-up, for a relatively small GRN, comprising of five genes. Keywords: Parallel SMBioNet, Gene Regulatory Network, Discrete Mod- eling 1 Introduction Living cells are complex power houses of human machinery, where the activities are organized as network of interacting entities such as Genes, mRNA and their products. Understanding the structural and behavioral orientation of cellular and sub-cellular interactions is fundamental step to uncover disease mechanics and realization of personalized medicine [4]. As the complexity of these networks increase, efficient computational methods are required not only to infer network information from gene expression data, but also to gain insight into regulatory mechanisms by finding answers to biological questions [6]. This growing complex- ity of GRNs also requires correlation mechanisms between experimental data, theory and computational methods [3]. Once a suitable framework is selected, and the system under investigation has been modeled, identification of parame- ters, coherent with biological knowledge, is a key problem known as Parameter Estimation or Network Inference [5]. These parameters are not a priori known, and have to be reverse engineered and compared with biological knowledge. The problem of parameter estimation is extremely important in systems biology and Proceedings IWBBIO 2014. Granada 7-9 April, 2014 1769