• Journal of Institute of Control, Robotics and Systems
• /
• v.10 no.12
• /
• pp.1189-1195
• /
• 2004

This paper investigates construction of gene (interaction) networks from gene expression time-series data based on evolutionary computation. To illustrate the proposed approach in a comprehensive way, we first assume an artificial gene network and then compare it with the reconstructed network from the gene expression time-series data generated by the artificial network. Next, we employ real gene expression time-series data (Spellman's yeast data) to construct a gene network by applying the proposed approach. From these experiments, we find that the proposed approach can be used as a useful tool for discovering the structure of a gene network as well as the corresponding relations among genes. The constructed gene network can further provide biologists with information to generate/test new hypotheses and ultimately to unravel the gene functions.

• Bioinformatics and Biosystems
• /
• v.2 no.1
• /
• pp.12-16
• /
• 2007

DNA repair means a collection of processes that a cell identifies and corrects damage to genome sequence. The DNA repair processes are important because a genome would not be able to maintain its essential cellular functions without the processes. In this research, we make some gene regulatory networks of DNA repair in S. cerevisiae to know how each gene interacts with others. Two approaches are adapted to make the networks; Bayesian Network and ARACNE. After construction of gene regulatory networks based on the two approaches, the two networks are compared to each other to predict which genes have important roles in the DNA repair processes by finding conserved interactions and looking for hubs. In addition, each interaction between genes in the networks is validated with interaction information in S. cerevisiae genome database to support the meaning of predicted interactions in the networks.

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2005.09a
• /
• pp.339-343
• /
• 2005

Boolean networks(BN) construction is one of the commonly used methods for building gene networks from time series microarray data. However, BN has two major drawbacks. First, it requires heavy computing times. Second, the binary transformation of the microarray data may cause a loss of information. This paper propose two methods using liner regression to construct gene regulatory networks. The first proposed method uses regression based BN variable selection method, which reduces the computing time significantly in the BN construction. The second method is the regression based network method that can flexibly incorporate the interaction of the genes using continuous gene expression data. We construct the network structure from the simulated data to compare the computing times between Boolean networks and the proposed method. The regression based network method is evaluated using a microarray data of cell cycle in Caulobacter crescentus.

• Interdisciplinary Bio Central
• /
• v.3 no.2
• /
• pp.8.1-8.6
• /
• 2011

Background: Thanks to the development of the mathematical/statistical reverse engineering and the high-throughput measuring biotechnology, lots of biologically meaningful genegene interaction networks have been revealed. Steady-state analysis of these systems provides an important clue to understand and to predict the systematic behaviours of the biological system. However, modeling such a complex and large-scale system is one of the challenging difficulties in systems biology. Results: We introduce a new stochastic modeling approach that can describe gene regulatory mechanisms by dividing two (DNA and protein) layers. Simple queuing system is employed to explain the DNA layer and the protein layer is modeled using G-networks which enable us to account for the post-translational protein interactions. Our method is applied to a transcription repression system and an active protein degradation system. The steady-state results suggest that the active protein degradation system is more sensitive but the transcription repression system might be more reliable than the transcription repression system. Conclusions: Our two layer stochastic model successfully describes the long-run behaviour of gene regulatory networks which consist of various mRNA/protein processes. The analytic solution of the G-networks enables us to extend our model to a large-scale system. A more reliable modeling approach could be achieved by cooperating with a real experimental study in synthetic biology.

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2003.10a
• /
• pp.164-169
• /
• 2003

In this paper, we propose the integrated Bayesian network framework to reconstruct genetic regulatory networks from genome expression data. The proposed model overcomes the dimensionality problem of multivariate analysis by building coherent sub-networks from confined gene clusters and combining these networks via intermediary points. Gene Shaving algorithm is used to cluster genes that share a common function or co-regulation. Retrieved clusters incorporate prior biological knowledge such as Gene Ontology, pathway, and protein protein interaction information for extracting other related genes. With these extended gene list, system builds genetic sub-networks using Bayesian network with MDL score and Sparse Candidate algorithm. Identifying functional modules of genes is done by not only microarray data itself but also well-proved biological knowledge. This integrated approach can improve there liability of a network in that false relations due to the lack of data can be reduced. Another advantage is the decreased computational complexity by constrained gene sets. To evaluate the proposed system, S. Cerevisiae cell cycle data [1] is applied. The result analysis presents new hypotheses about novel genetic interactions as well as typical relationships known by previous researches [2].

• Genomics & Informatics
• /
• v.11 no.4
• /
• pp.200-210
• /
• 2013

Studying biological networks, such as protein-protein interactions, is key to understanding complex biological activities. Various types of large-scale biological datasets have been collected and analyzed with high-throughput technologies, including DNA microarray, next-generation sequencing, and the two-hybrid screening system, for this purpose. In this review, we focus on network-based approaches that help in understanding biological systems and identifying biological functions. Accordingly, this paper covers two major topics in network biology: reconstruction of gene regulatory networks and network-based applications, including protein function prediction, disease gene prioritization, and network-based genome-wide association study.

• The KIPS Transactions:PartD
• /
• v.14D no.1 s.111
• /
• pp.9-20
• /
• 2007

Groups of genes control the functioning of a cell by complex interactions. Such interactions of gene groups are tailed Gene Regulatory Networks(GRNs). Two previous data mining approaches, clustering and classification, have been used to analyze gene expression data. Though these mining tools are useful for determining membership of genes by homology, they don't identify the regulatory relationships among genes found in the same class of molecular actions. Furthermore, we need to understand the mechanism of how genes relate and how they regulate one another. In order to detect regulatory relationships among genes from time-series Microarray data, we propose a novel approach using frequent pattern mining and chain rules. In this approach, we propose a method for transforming gene expression data to make suitable for frequent pattern mining, and gene expression patterns we detected by applying the FP-growth algorithm. Next, we construct a gene regulatory network from frequent gene patterns using chain rules. Finally, we validate our proposed method through our experimental results, which are consistent with published results.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2014.11a
• /
• pp.837-840
• /
• 2014

Interactions between genes have long been recognized and studied by many researchers, and they formed a large-scale interaction networks. In systems biology, it has been a challenge to investigate the factors to determine network dynamics. Here, we create a new network called an effectiveness network by calculating thy dynamical effectiveness from a node to another node. We found that robust nodes tend to have smaller number of edges than non-robust nodes. This implies that hub nodes are likely to affect the network robustness.

• Bioinformatics and Biosystems
• /
• v.1 no.1
• /
• pp.38-45
• /
• 2006

We present a client-server system architecture for inferring genetic interaction networks based on Bayesian networks. It is typical to take tens of hours when genome-wide large-scale genetic interaction networks are inferred in the form of Bayesian networks. To deal with this situation, batch-style distributed system architectures are preferable to interactive standalone architectures. Thus, we have implemented a loosely coupled client-server system for network inference and user interface. The network inference consists of two stages. Firstly, the proposed method divides a whole gene set into overlapped modules, based on biological annotations and expression data together. Secondly, it infers Bayesian networks for each module, and integrates the learned subnetworks to a global network through common genes across the modules.

• Asian Pacific Journal of Cancer Prevention
• /
• v.14 no.3
• /
• pp.1731-1735
• /
• 2013

Objective: We aimed to identify key genes, pathways and function modules in the development of diffuse large B-cell lymphoma (DLBCL) with microarray data and interaction network analysis. Methods: Microarray data sets for 7 DLBCL samples and 7 normal controls was downloaded from the Gene Expression Omnibus (GEO) database and differentially expressed genes (DEGs) were identified with Student's t-test. KEGG functional enrichment analysis was performed to uncover their biological functions. Three global networks were established for immune system, signaling molecules and interactions and cancer genes. The DEGs were compared with the networks to observe their distributions and determine important key genes, pathways and modules. Results: A total of 945 DEGs were obtained, 272 up-regulated and 673 down-regulated. KEGG analysis revealed that two groups of pathways were significantly enriched: immune function and signaling molecules and interactions. Following interaction network analysis further confirmed the association of DEGs in immune system, signaling molecules and interactions and cancer genes. Conclusions: Our study could systemically characterize gene expression changes in DLBCL with microarray technology. A range of key genes, pathways and function modules were revealed. Utility in diagnosis and treatment may be expected with further focused research.