Show simple item record

dc.contributor.author
Roth, Volker
dc.contributor.author
Fischer, Bernd
dc.date.accessioned
2018-09-04T14:22:19Z
dc.date.available
2017-06-08T18:05:19Z
dc.date.available
2018-09-04T14:22:19Z
dc.date.issued
2007-05
dc.identifier.issn
1471-2105
dc.identifier.other
10.1186/1471-2105-8-S2-S12
en_US
dc.identifier.uri
http://hdl.handle.net/20.500.11850/8235
dc.identifier.doi
10.3929/ethz-b-000008235
dc.description.abstract
Background We develop a probabilistic model for combining kernel matrices to predict the function of proteins. It extends previous approaches in that it can handle multiple labels which naturally appear in the context of protein function. Results Explicit modeling of multilabels significantly improves the capability of learning protein function from multiple kernels. The performance and the interpretability of the inference model are further improved by simultaneously predicting the subcellular localization of proteins and by combining pairwise classifiers to consistent class membership estimates. Conclusion For the purpose of functional prediction of proteins, multilabels provide valuable information that should be included adequately in the training process of classifiers. Learning of functional categories gains from co-prediction of subcellular localization. Pairwise separation rules allow very detailed insights into the relevance of different measurements like sequence, structure, interaction data, or expression data. A preliminary version of the software can be downloaded from http://www.inf.ethz.ch/personal/vroth/KernelHMM/.
en_US
dc.format
application/pdf
en_US
dc.language.iso
en
en_US
dc.publisher
BioMed Central
en_US
dc.rights.uri
http://creativecommons.org/licenses/by/2.0/
dc.subject
Hide Markov Model
en_US
dc.subject
Linear Discriminant Analysis
en_US
dc.subject
Gaussian Mixture Model
en_US
dc.subject
Diffusion Kernel
en_US
dc.subject
Membership Probability
en_US
dc.title
Improved functional prediction of proteins by learning kernel combinations in multilabel settings
en_US
dc.type
Conference Paper
dc.rights.license
Creative Commons Attribution 2.0 Generic
ethz.journal.title
BMC Bioinformatics
ethz.journal.volume
8
en_US
ethz.journal.issue
Supplement 2
en_US
ethz.journal.abbreviated
BMC bioinformatics
ethz.pages.start
S12
en_US
ethz.size
12 p.
en_US
ethz.version.deposit
publishedVersion
en_US
ethz.event
2006 Conference on Probabilistic Modeling and Machine Learning in Structural and Systems Biology (PMSB)
en_US
ethz.event.location
Tuusula, Finland
en_US
ethz.event.date
June 17-18, 2006
en_US
ethz.identifier.nebis
004240301
ethz.publication.place
London
en_US
ethz.publication.status
published
en_US
ethz.leitzahl
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02150 - Dep. Informatik / Dep. of Computer Science::02661 - Institut für Maschinelles Lernen / Institute for Machine Learning::03659 - Buhmann, Joachim M. / Buhmann, Joachim M.
en_US
ethz.leitzahl.certified
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02150 - Dep. Informatik / Dep. of Computer Science::02661 - Institut für Maschinelles Lernen / Institute for Machine Learning::03659 - Buhmann, Joachim M. / Buhmann, Joachim M.
ethz.date.deposited
2017-06-08T18:05:35Z
ethz.source
ECIT
ethz.identifier.importid
imp59364bc35488d65507
ethz.ecitpid
pub:18943
ethz.eth
yes
en_US
ethz.availability
Open access
en_US
ethz.rosetta.installDate
2017-08-03T10:50:36Z
ethz.rosetta.lastUpdated
2018-11-08T01:34:40Z
ethz.rosetta.versionExported
true
ethz.COinS
ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.atitle=Improved%20functional%20prediction%20of%20proteins%20by%20learning%20kernel%20combinations%20in%20multilabel%20settings&rft.jtitle=BMC%20Bioinformatics&rft.date=2007-05&rft.volume=8&rft.issue=Supplement%202&rft.spage=S12&rft.issn=1471-2105&rft.au=Roth,%20Volker&Fischer,%20Bernd&rft.genre=proceeding&
 Search via SFX

Files in this item

Thumbnail

Publication type

Show simple item record