Analysis and Research of Key Genes in Gene Expression Network Based on Complex Network

Guobin Chen,Chao Tang,Ying Wang,Jun Qi,Yongzhong Wu,Xiaolong Shi,Atif Khan

doi:10.1155/2020/8816633

Guobin Chen, Chao Tang + Show 5 more

Open Access

https://doi.org/10.1155/2020/8816633

Copy DOI

Abstract

Gene expression network is also a type of complex network. It is challenging to analyze the gene expression network through relevant knowledge and algorithms of a complex network. In this paper, the existing characteristics of genes are analyzed from various indexes of the gene expression network to analyze key genes and TOP genes. Firstly, gene chip data are screened, gene data with obvious characteristics are selected, and relevant clustering characteristics are analyzed. Then, the complex gene network structure is established, and gene networks with different threshold shapes and different sizes are selected. Finally, the relevant indexes and PR values after the PageRank algorithm are analyzed for complex networks under different thresholds, thus establishing the TOP gene and PR sequence.

Highlights

With the development of gene chip and second-generation sequencing and the emergence of new technologies, the analysis of the human genetic structure has become a reality, the association information between genes is analyzed, being able to express genes related to key genes, and it is possible to analyze the correlation between genes and diseases more effectively
Unlike the research on inherent single molecules, the information at the whole system level can be displayed by establishing networks
In order to expand the scope and depth of the current research, the transition from a single molecule to a network system level becomes the major development direction of interactive networks because of, in general, the complex biological phenomena and pathology that cannot be caused by only one factor. erefore, as a tool to explore complex pathological and life phenomena, the interactive network provides an effective analysis method, to perfect the comprehensive research. ese networks have great research value, among them; the gene coexpression network has irreplaceable research characteristics in some aspects

Summary

Introduction

With the development of gene chip and second-generation sequencing and the emergence of new technologies, the analysis of the human genetic structure has become a reality, the association information between genes is analyzed, being able to express genes related to key genes, and it is possible to analyze the correlation between genes and diseases more effectively. Based on the supervised learning method, we mainly use the known rules to infer gene regulatory networks on genome-wide data, such as SEREND [14], GENIES [15], and SIRENE [18], but all need additional information of regulatory relationships to cultivate models. Constructing complex gene expression networks and using large-scale gene expression data sets for network analysis are effective methods to reveal new biological knowledge. Hua et al [20] proposed the fusion research of three commonly used reasoning algorithms to establish a genome-scale and high-quality gene coexpression network. After applying this expression network to monocotyledonous plant rice, the network quality has been verified and evaluated through the selected gene function association data sets, which is obviously superior to other methods. In order to evaluate the efficiency of the proposed method, the commonly used classical clustering algorithm has obvious advantages in identifying protein complexes in ST-APIN and the other three dynamic PIN

Technical Indicators Related to Complex Networks

Key Gene Determination Method Based on PageRank Algorithm

Experimental Analysis

Conclusion