File Download
 
Links for fulltext
(May Require Subscription)
 
Supplementary

Article: MotifVoter: A novel ensemble method for fine-grained integration of generic motif finders
  • Basic View
  • Metadata View
  • XML View
TitleMotifVoter: A novel ensemble method for fine-grained integration of generic motif finders
 
AuthorsWijaya, E2 4
Yiu, SM1
Son, NT4
Kanagasabai, R2
Sung, WK4 3
 
Issue Date2008
 
PublisherOxford University Press. The Journal's web site is located at http://bioinformatics.oxfordjournals.org/
 
CitationBioinformatics, 2008, v. 24 n. 20, p. 2288-2295 [How to Cite?]
DOI: http://dx.doi.org/10.1093/bioinformatics/btn420
 
AbstractMotivation: Locating transcription factor binding sites (motifs) is a key step in understanding gene regulation. Based on Tompa's benchmark study, the performance of current de novo motif finders is far from satisfactory (with sensitivity ≤0.222 and precision ≤0.307). The same study also shows that no motif finder performs consistently well over all datasets. Hence, it is not clear which finder one should use for a given dataset. To address this issue, a class of algorithms called ensemble methods have been proposed. Though the existing ensemble methods overall perform better than stand-alone motif finders, the improvement gained is not substantial. Our study reveals that these methods do not fully exploit the information obtained from the results of individual finders, resulting in minor improvement in sensitivity and poor precision. Results: In this article, we identify several key observations on how to utilize the results from individual finders and design a novel ensemble method, MotifVoter, to predict the motifs and binding sites. Evaluations on 186 datasets show that MotifVoter can locate more than 95% of the binding sites found by its component motif finders. In terms of sensitivity and precision, MotifVoter outperforms stand-alone motif finders and ensemble methods significantly on Tompa's benchmark, Escherichia coli, and ChIP-Chip datasets. MotifVoter is available online via a web server with several biologist-friendly features. © The Author 2008. Published by Oxford University Press. All rights reserved.
 
ISSN1367-4803
2013 Impact Factor: 4.621
 
DOIhttp://dx.doi.org/10.1093/bioinformatics/btn420
 
ISI Accession Number IDWOS:000259973500003
Funding AgencyGrant Number
National University of SingaporeR-252-000326-112
Research Output Prize (Faculty of Engineering) of the University of HongKong
Funding Information:

National University of Singapore (grant R-252-000326-112); Research Output Prize (Faculty of Engineering) of the University of HongKong to S.M.Y.

 
ReferencesReferences in Scopus
 
DC FieldValue
dc.contributor.authorWijaya, E
 
dc.contributor.authorYiu, SM
 
dc.contributor.authorSon, NT
 
dc.contributor.authorKanagasabai, R
 
dc.contributor.authorSung, WK
 
dc.date.accessioned2010-05-31T04:14:45Z
 
dc.date.available2010-05-31T04:14:45Z
 
dc.date.issued2008
 
dc.description.abstractMotivation: Locating transcription factor binding sites (motifs) is a key step in understanding gene regulation. Based on Tompa's benchmark study, the performance of current de novo motif finders is far from satisfactory (with sensitivity ≤0.222 and precision ≤0.307). The same study also shows that no motif finder performs consistently well over all datasets. Hence, it is not clear which finder one should use for a given dataset. To address this issue, a class of algorithms called ensemble methods have been proposed. Though the existing ensemble methods overall perform better than stand-alone motif finders, the improvement gained is not substantial. Our study reveals that these methods do not fully exploit the information obtained from the results of individual finders, resulting in minor improvement in sensitivity and poor precision. Results: In this article, we identify several key observations on how to utilize the results from individual finders and design a novel ensemble method, MotifVoter, to predict the motifs and binding sites. Evaluations on 186 datasets show that MotifVoter can locate more than 95% of the binding sites found by its component motif finders. In terms of sensitivity and precision, MotifVoter outperforms stand-alone motif finders and ensemble methods significantly on Tompa's benchmark, Escherichia coli, and ChIP-Chip datasets. MotifVoter is available online via a web server with several biologist-friendly features. © The Author 2008. Published by Oxford University Press. All rights reserved.
 
dc.description.naturelink_to_OA_fulltext
 
dc.identifier.citationBioinformatics, 2008, v. 24 n. 20, p. 2288-2295 [How to Cite?]
DOI: http://dx.doi.org/10.1093/bioinformatics/btn420
 
dc.identifier.citeulike3132697
 
dc.identifier.doihttp://dx.doi.org/10.1093/bioinformatics/btn420
 
dc.identifier.eissn1460-2059
 
dc.identifier.epage2295
 
dc.identifier.hkuros161318
 
dc.identifier.isiWOS:000259973500003
Funding AgencyGrant Number
National University of SingaporeR-252-000326-112
Research Output Prize (Faculty of Engineering) of the University of HongKong
Funding Information:

National University of Singapore (grant R-252-000326-112); Research Output Prize (Faculty of Engineering) of the University of HongKong to S.M.Y.

 
dc.identifier.issn1367-4803
2013 Impact Factor: 4.621
 
dc.identifier.issue20
 
dc.identifier.openurl
 
dc.identifier.pmid18697768
 
dc.identifier.scopuseid_2-s2.0-53749085875
 
dc.identifier.spage2288
 
dc.identifier.urihttp://hdl.handle.net/10722/60600
 
dc.identifier.volume24
 
dc.languageeng
 
dc.publisherOxford University Press. The Journal's web site is located at http://bioinformatics.oxfordjournals.org/
 
dc.publisher.placeUnited Kingdom
 
dc.relation.ispartofBioinformatics
 
dc.relation.referencesReferences in Scopus
 
dc.rightsBioinformatics. Copyright © Oxford University Press.
 
dc.subject.meshComputational Biology - methods
 
dc.subject.meshRegulatory Elements, Transcriptional
 
dc.subject.meshTranscription Factors - chemistry - metabolism
 
dc.subject.meshProtein Structure, Tertiary
 
dc.titleMotifVoter: A novel ensemble method for fine-grained integration of generic motif finders
 
dc.typeArticle
 
<?xml encoding="utf-8" version="1.0"?>
<item><contributor.author>Wijaya, E</contributor.author>
<contributor.author>Yiu, SM</contributor.author>
<contributor.author>Son, NT</contributor.author>
<contributor.author>Kanagasabai, R</contributor.author>
<contributor.author>Sung, WK</contributor.author>
<date.accessioned>2010-05-31T04:14:45Z</date.accessioned>
<date.available>2010-05-31T04:14:45Z</date.available>
<date.issued>2008</date.issued>
<identifier.citation>Bioinformatics, 2008, v. 24 n. 20, p. 2288-2295</identifier.citation>
<identifier.issn>1367-4803</identifier.issn>
<identifier.uri>http://hdl.handle.net/10722/60600</identifier.uri>
<description.abstract>Motivation: Locating transcription factor binding sites (motifs) is a key step in understanding gene regulation. Based on Tompa&apos;s benchmark study, the performance of current de novo motif finders is far from satisfactory (with sensitivity &#8804;0.222 and precision &#8804;0.307). The same study also shows that no motif finder performs consistently well over all datasets. Hence, it is not clear which finder one should use for a given dataset. To address this issue, a class of algorithms called ensemble methods have been proposed. Though the existing ensemble methods overall perform better than stand-alone motif finders, the improvement gained is not substantial. Our study reveals that these methods do not fully exploit the information obtained from the results of individual finders, resulting in minor improvement in sensitivity and poor precision. Results: In this article, we identify several key observations on how to utilize the results from individual finders and design a novel ensemble method, MotifVoter, to predict the motifs and binding sites. Evaluations on 186 datasets show that MotifVoter can locate more than 95% of the binding sites found by its component motif finders. In terms of sensitivity and precision, MotifVoter outperforms stand-alone motif finders and ensemble methods significantly on Tompa&apos;s benchmark, Escherichia coli, and ChIP-Chip datasets. MotifVoter is available online via a web server with several biologist-friendly features. &#169; The Author 2008. Published by Oxford University Press. All rights reserved.</description.abstract>
<language>eng</language>
<publisher>Oxford University Press. The Journal&apos;s web site is located at http://bioinformatics.oxfordjournals.org/</publisher>
<relation.ispartof>Bioinformatics</relation.ispartof>
<rights>Bioinformatics. Copyright &#169; Oxford University Press.</rights>
<subject.mesh>Computational Biology - methods</subject.mesh>
<subject.mesh>Regulatory Elements, Transcriptional</subject.mesh>
<subject.mesh>Transcription Factors - chemistry - metabolism</subject.mesh>
<subject.mesh>Protein Structure, Tertiary</subject.mesh>
<title>MotifVoter: A novel ensemble method for fine-grained integration of generic motif finders</title>
<type>Article</type>
<identifier.openurl>http://library.hku.hk:4550/resserv?sid=HKU:IR&amp;issn=1367-4803&amp;volume=24&amp;issue=20&amp;spage=2288&amp;epage=2295&amp;date=2008&amp;atitle=MotifVoter:+a+novel+ensemble+method+for+fine-grained+integration+of+generic+motif+finders</identifier.openurl>
<description.nature>link_to_OA_fulltext</description.nature>
<identifier.doi>10.1093/bioinformatics/btn420</identifier.doi>
<identifier.pmid>18697768</identifier.pmid>
<identifier.scopus>eid_2-s2.0-53749085875</identifier.scopus>
<identifier.hkuros>161318</identifier.hkuros>
<relation.references>http://www.scopus.com/mlt/select.url?eid=2-s2.0-53749085875&amp;selection=ref&amp;src=s&amp;origin=recordpage</relation.references>
<identifier.volume>24</identifier.volume>
<identifier.issue>20</identifier.issue>
<identifier.spage>2288</identifier.spage>
<identifier.epage>2295</identifier.epage>
<identifier.eissn>1460-2059</identifier.eissn>
<identifier.isi>WOS:000259973500003</identifier.isi>
<publisher.place>United Kingdom</publisher.place>
<identifier.citeulike>3132697</identifier.citeulike>
<bitstream.url>http://hub.hku.hk/bitstream/10722/60600/1/re01.html</bitstream.url>
</item>
Author Affiliations
  1. The University of Hong Kong
  2. Institute for Infocomm Research, A-Star, Singapore
  3. Genome Institute of Singapore
  4. National University of Singapore