A Hybrid Approach for CpG Island Detection in the Human Genome.

BACKGROUND:CpG islands have been demonstrated to influence local chromatin structures and simplify the regulation of gene activity. However, the accurate and rapid determination of CpG islands for whole DNA sequences remains experimentally and computationally challenging. METHODOLOGY/PRINCIPAL FINDI...

Full description

Bibliographic Details
Main Authors: Cheng-Hong Yang, Yu-Da Lin, Yi-Cheng Chiang, Li-Yeh Chuang
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2016-01-01
Series:PLoS ONE
Online Access:http://europepmc.org/articles/PMC4705099?pdf=render
_version_ 1818033573394907136
author Cheng-Hong Yang
Yu-Da Lin
Yi-Cheng Chiang
Li-Yeh Chuang
author_facet Cheng-Hong Yang
Yu-Da Lin
Yi-Cheng Chiang
Li-Yeh Chuang
author_sort Cheng-Hong Yang
collection DOAJ
description BACKGROUND:CpG islands have been demonstrated to influence local chromatin structures and simplify the regulation of gene activity. However, the accurate and rapid determination of CpG islands for whole DNA sequences remains experimentally and computationally challenging. METHODOLOGY/PRINCIPAL FINDINGS:A novel procedure is proposed to detect CpG islands by combining clustering technology with the sliding-window method (PSO-based). Clustering technology is used to detect the locations of all possible CpG islands and process the data, thus effectively obviating the need for the extensive and unnecessary processing of DNA fragments, and thus improving the efficiency of sliding-window based particle swarm optimization (PSO) search. This proposed approach, named ClusterPSO, provides versatile and highly-sensitive detection of CpG islands in the human genome. In addition, the detection efficiency of ClusterPSO is compared with eight CpG island detection methods in the human genome. Comparison of the detection efficiency for the CpG islands in human genome, including sensitivity, specificity, accuracy, performance coefficient (PC), and correlation coefficient (CC), ClusterPSO revealed superior detection ability among all of the test methods. Moreover, the combination of clustering technology and PSO method can successfully overcome their respective drawbacks while maintaining their advantages. Thus, clustering technology could be hybridized with the optimization algorithm method to optimize CpG island detection. CONCLUSION/SIGNIFICANCE:The prediction accuracy of ClusterPSO was quite high, indicating the combination of CpGcluster and PSO has several advantages over CpGcluster and PSO alone. In addition, ClusterPSO significantly reduced implementation time.
first_indexed 2024-12-10T06:25:25Z
format Article
id doaj.art-3657f9691ce74d5d81a68e183147f19f
institution Directory Open Access Journal
issn 1932-6203
language English
last_indexed 2024-12-10T06:25:25Z
publishDate 2016-01-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS ONE
spelling doaj.art-3657f9691ce74d5d81a68e183147f19f2022-12-22T01:59:13ZengPublic Library of Science (PLoS)PLoS ONE1932-62032016-01-01111e014474810.1371/journal.pone.0144748A Hybrid Approach for CpG Island Detection in the Human Genome.Cheng-Hong YangYu-Da LinYi-Cheng ChiangLi-Yeh ChuangBACKGROUND:CpG islands have been demonstrated to influence local chromatin structures and simplify the regulation of gene activity. However, the accurate and rapid determination of CpG islands for whole DNA sequences remains experimentally and computationally challenging. METHODOLOGY/PRINCIPAL FINDINGS:A novel procedure is proposed to detect CpG islands by combining clustering technology with the sliding-window method (PSO-based). Clustering technology is used to detect the locations of all possible CpG islands and process the data, thus effectively obviating the need for the extensive and unnecessary processing of DNA fragments, and thus improving the efficiency of sliding-window based particle swarm optimization (PSO) search. This proposed approach, named ClusterPSO, provides versatile and highly-sensitive detection of CpG islands in the human genome. In addition, the detection efficiency of ClusterPSO is compared with eight CpG island detection methods in the human genome. Comparison of the detection efficiency for the CpG islands in human genome, including sensitivity, specificity, accuracy, performance coefficient (PC), and correlation coefficient (CC), ClusterPSO revealed superior detection ability among all of the test methods. Moreover, the combination of clustering technology and PSO method can successfully overcome their respective drawbacks while maintaining their advantages. Thus, clustering technology could be hybridized with the optimization algorithm method to optimize CpG island detection. CONCLUSION/SIGNIFICANCE:The prediction accuracy of ClusterPSO was quite high, indicating the combination of CpGcluster and PSO has several advantages over CpGcluster and PSO alone. In addition, ClusterPSO significantly reduced implementation time.http://europepmc.org/articles/PMC4705099?pdf=render
spellingShingle Cheng-Hong Yang
Yu-Da Lin
Yi-Cheng Chiang
Li-Yeh Chuang
A Hybrid Approach for CpG Island Detection in the Human Genome.
PLoS ONE
title A Hybrid Approach for CpG Island Detection in the Human Genome.
title_full A Hybrid Approach for CpG Island Detection in the Human Genome.
title_fullStr A Hybrid Approach for CpG Island Detection in the Human Genome.
title_full_unstemmed A Hybrid Approach for CpG Island Detection in the Human Genome.
title_short A Hybrid Approach for CpG Island Detection in the Human Genome.
title_sort hybrid approach for cpg island detection in the human genome
url http://europepmc.org/articles/PMC4705099?pdf=render
work_keys_str_mv AT chenghongyang ahybridapproachforcpgislanddetectioninthehumangenome
AT yudalin ahybridapproachforcpgislanddetectioninthehumangenome
AT yichengchiang ahybridapproachforcpgislanddetectioninthehumangenome
AT liyehchuang ahybridapproachforcpgislanddetectioninthehumangenome
AT chenghongyang hybridapproachforcpgislanddetectioninthehumangenome
AT yudalin hybridapproachforcpgislanddetectioninthehumangenome
AT yichengchiang hybridapproachforcpgislanddetectioninthehumangenome
AT liyehchuang hybridapproachforcpgislanddetectioninthehumangenome