Reconstruction and analysis of gene regulatory network for lung cancer using relational matrix and neuro-fuzzy system
- Authors
- Shin, B.; Wang, B.; Lim, J.S.
- Issue Date
- Aug-2017
- Publisher
- Institute of Advanced Scientific Research, Inc.
- Keywords
- Gene Regulatory Network; Lung Cancer; Relational Matrix; System Biology; Weighted Neuro Fuzzy Algorithm
- Citation
- Journal of Advanced Research in Dynamical and Control Systems, v.9, no.8, pp.157 - 163
- Journal Title
- Journal of Advanced Research in Dynamical and Control Systems
- Volume
- 9
- Number
- 8
- Start Page
- 157
- End Page
- 163
- URI
- https://scholarworks.bwise.kr/gachon/handle/2020.sw.gachon/80288
- ISSN
- 1943-023X
- Abstract
- Objectives: Owing to the appearance of microarray technology, researchers were able to obtain huge amounts of gene expression data. This vast amount of microarray data and computational methods has allowed researchers to look at overall biological mechanisms as a network. Gene regulatory network which models interactions among genes from microarray data helps to reveal pathways associated with disease and predict the effect of the drug. Methods/Statistical analysis: In this paper, a relational matrix which is used to reconstruct the gene regulatory network of lung cancer is proposed, and we demonstrate that by weighted neuro fuzzy algorithm, the groups of interacting genes founded by scanning relational matrix can classify the types of lung cancer very well. The relational matrix is constructed by counting the number of meaningful relationship between two genes for all samples. The weighted neuro fuzzy algorithm usesa bounded sum function that the three functions are combined into one for learning and classifying. Findings: We discovered more than about 500 genes which have strong relationships with the one or more genes by investigating the matrix after constructing the proposed relational matrix, and reconstructed gene regulatory network with the discovered genes. We were able to find that only 42 genes in reconstructed gene regulatory network receive affect from a large number of genes. We were also able to find that they were connected with each other. When we investigated the other target genes which have a target geneas regulator, it played a role as regulator in almost all 42 target genes. From this result, we can know that there exist a group which is consisted of genes that have very high interactions in gene regulatory network, and the genes included in the group play an important role for lung cancer. We discovered the genes which have high relationship according to the type of lung cancer. These genes were used for classifying the type of lung cancer. Improvements/Applications: When the types of lung cancer were classified through machine learning by the weighted neuro fuzzy algorithm with those genes, we got the accuracy of 99.5074% for classifying five types of genes. © 2017, Institute of Advanced Scientific Research, Inc.. All rights reserved.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - IT융합대학 > 컴퓨터공학과 > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/gachon/handle/2020.sw.gachon/80288)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.