MODELLING OF GENE REGULATORY NETWORK RECONSTRUCTION PROCEDURE BASED ON THE COMPLEX USE OF TOPOLOGICAL PARAMETERS
DOI:
https://doi.org/10.32782/KNTU2618-0340/2021.4.1.3Keywords:
gene regulatory network, network topology, topological parameters, Harrington desirability function, thresholding, correlation inference algorithmAbstract
The paper presents the simulation results concerning the determination of the gene regulatory network optimal topology during its reconstruction using the correlation inference algorithm. The gene regulatory network was presented as an undirected graph, in which the nodes are genes or metabolites, and the arcs define the connection between the corresponding network elements. The assessment of the network topology was carried out by calculating the values of single topological parameters, taking into account both the structure of the network and the nature of the connections between the corresponding elements. The following single topological parameters were investigated: the number of nodes in the network, the degree of nodes or their connectivity, the density of the network, the coefficients of clustering and centralization, and network heterogeneity. The final decision regarding the network topology was dine based on the analysis of the generalized topological index, which was calculated using the Harrington desirability function. Modelling of the gene network reconstruction process based on gene expression profiles was carried out in the Cytoscape software environment using the moe430 gene expression profiles of the ArrayExpress database, which contains information concerning the genes expression of two types of mesenchymal cells: neural crest and mesoderm. The process of gene regulatory network reconstruction was carried out using a correlation inference algorithm, the practical implementation of which involves calculating the pair correlation coefficients between the studied gene expression profiles. The network topology, in this case, was formed on the basis of the thresholding coefficient τ, which determines the threshold value for the presence of a connection between a pair of corresponding network genes. As a result of modelling, the diagrams of the distribution of single topological parameters and the generalized topological index versus the value of the thresholding coefficient were created. The analysis of these diagrams can allow us to determine the gene regulatory network optimal topology.
References
D‘haeseleer P., Wen X., Fuhrman S., Somogyi R. Linear modeling of mRNA expression levels during CNS development and injury. Pacific Symposium on Biocomputing. 1999. Р. 41–52.
Liang S., Fuhrman S., Somogyi R. REVEAL, a general reverse engineering algorithm for inference of genetic network architectures. Pacific Symposium on Biocomp. 1998. Р. 18–29.
Friedman N., Linial M., Nachman I., Pe'er D. Using Bayesian networks to analyse expression data. Journal of Computational Biology. 2000. №7(3-4). Р. 601–620.
Chen T., He H. L., Church G. M. Modeling gene expression with differential equations. Proceedings of the Pacific Symposium on Biocomputing. 1999. Р. 29–40.
Bansal M., Belcastro V., Ambesi-Impiombato A., di Bernardo D. How to infer gene networks from expression profiles. Molecular Systems Biology. 2007. №3(78). Р.1–10.
Yaghoobi H., Haghipour S., Hamzeiy H., et al. A Review of Modeling Techniques for Genetic Regulatory Networks. Journal of Medical Signals and Sensors. 2012. №2(1). Р.61–70.
Emmert-Streib F., Dehmer M., Haibe-Kains B. Gene regulatory networks and their applications: understanding biological and medical problems in terms of networks. Frontiers in cell and developmental biology. 2014. Р. 1–7.
Soneson C., Delorenzi M. A comparison of methods for differential expression analysis of RNA-seq data. BMC Bioinformatics. 2013. №14(91). Р. 1–18.
Arnone M.I., Davidson E.H. The hardwiring of development: organization and function of genomic regulatory systems. Development. 1997. №124(10). Р. 1851–1864.
Jeong H., Tombor B., Albert R., et al. The large-scale organization of metabolic networks. Nature. 2000. № 407(6804). Р. 651–654.
Bork P., Jensen L.J., von Mering C., et al. Protein interaction networks from yeast to human. Current Opinion in Structural Biology. 2004. №14(3). Р. 292–299.
Babichev S., Taif M. A, Lytvynenko V. Inductive model of data clustering based on the agglomerative hierarchical algorithm. Proceeding of the 2016 IEEE First International Conference on Data Stream Mining and Processing (DSMP). 2016. Р.19–22.
Babichev S., Taif M. A., Lytvynenko V., Korobchynskyi M. Objective clustering inductive technology of gene expression sequences features. Communication in Computer and Information Science. 2017. № 716. Р. 359–372.
Babichev S., Lytvynenko V., Skvor J., Fiser J. Model of the objective clustering inductive technology of gene expression profiles based on SOTA and DBSCAN clustering algorithms. Advances in Intelligent Systems and Computing. 2018. №.689. Р.21–39.
Pontes B., Giráldez R., Aguilar-Ruiz J. S. Biclustering on expression data: A review. Journal of Biomedical Informatics. 2015. № 57. Р. 163–180.
Assenov Y., Ramırez F., Schelhorn S.-E., et al. Computing topological parameters of biological networks. Bioinformatics. 2008. №24(2). Р. 282–284.
Shannon P., Markiel A., et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003. P. 2498–2504.
Bhattacherjee V., Mukhopadhyay P., et al. Neural crest and mesoderm lineagedependent gene expression in orofacial development. Differentiation, 2007. P. 128–139.
