File Download
  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Collective prediction of protein functions from protein-protein interaction networks

TitleCollective prediction of protein functions from protein-protein interaction networks
Authors
Issue Date2014
Citation
BMC Bioinformatics, 2014, v. 15, suppl. 2, article no. S9 How to Cite?
Abstract© 2014 Wu et al.; licensee BioMed Central Ltd. Background: Automated assignment of functions to unknown proteins is one of the most important task in computational biology. The development of experimental methods for genome scale analysis of molecular interaction networks offers new ways to infer protein function from protein-protein interaction (PPI) network data. Existing techniques for collective classification (CC) usually increase accuracy for network data, wherein instances are interlinked with each other, using a large amount of labeled data for training. However, the labeled data are timeconsuming and expensive to obtain. On the other hand, one can easily obtain large amount of unlabeled data. Thus, more sophisticated methods are needed to exploit the unlabeled data to increase prediction accuracy for protein function prediction. Results: In this paper, we propose an effective Markov chain based CC algorithm (ICAM) to tackle the label deficiency problem in CC for interrelated proteins from PPI networks. Our idea is to model the problem using two distinct Markov chain classifiers to make separate predictions with regard to attribute features from protein data and relational features from relational information. The ICAM learning algorithm combines the results of the two classifiers to compute the ranks of labels to indicate the importance of a set of labels to an instance, and uses an ICA framework to iteratively refine the learning models for improving performance of protein function prediction from PPI networks in the paucity of labeled data. Conclusion: Experimental results on the real-world Yeast protein-protein interaction datasets show that our proposed ICAM method is better than the other ICA-type methods given limited labeled training data. This approach can serve as a valuable tool for the study of protein function prediction from PPI networks.
Persistent Identifierhttp://hdl.handle.net/10722/276682
ISSN
2017 Impact Factor: 2.213
2015 SCImago Journal Rankings: 1.722
PubMed Central ID

 

DC FieldValueLanguage
dc.contributor.authorWu, Qingyao-
dc.contributor.authorYe, Yunming-
dc.contributor.authorNg, Michael K.-
dc.contributor.authorHo, Shen Shyang-
dc.contributor.authorShi, Ruichao-
dc.date.accessioned2019-09-18T08:34:21Z-
dc.date.available2019-09-18T08:34:21Z-
dc.date.issued2014-
dc.identifier.citationBMC Bioinformatics, 2014, v. 15, suppl. 2, article no. S9-
dc.identifier.issn1471-2105-
dc.identifier.urihttp://hdl.handle.net/10722/276682-
dc.description.abstract© 2014 Wu et al.; licensee BioMed Central Ltd. Background: Automated assignment of functions to unknown proteins is one of the most important task in computational biology. The development of experimental methods for genome scale analysis of molecular interaction networks offers new ways to infer protein function from protein-protein interaction (PPI) network data. Existing techniques for collective classification (CC) usually increase accuracy for network data, wherein instances are interlinked with each other, using a large amount of labeled data for training. However, the labeled data are timeconsuming and expensive to obtain. On the other hand, one can easily obtain large amount of unlabeled data. Thus, more sophisticated methods are needed to exploit the unlabeled data to increase prediction accuracy for protein function prediction. Results: In this paper, we propose an effective Markov chain based CC algorithm (ICAM) to tackle the label deficiency problem in CC for interrelated proteins from PPI networks. Our idea is to model the problem using two distinct Markov chain classifiers to make separate predictions with regard to attribute features from protein data and relational features from relational information. The ICAM learning algorithm combines the results of the two classifiers to compute the ranks of labels to indicate the importance of a set of labels to an instance, and uses an ICA framework to iteratively refine the learning models for improving performance of protein function prediction from PPI networks in the paucity of labeled data. Conclusion: Experimental results on the real-world Yeast protein-protein interaction datasets show that our proposed ICAM method is better than the other ICA-type methods given limited labeled training data. This approach can serve as a valuable tool for the study of protein function prediction from PPI networks.-
dc.languageeng-
dc.relation.ispartofBMC Bioinformatics-
dc.rightsThis work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.-
dc.titleCollective prediction of protein functions from protein-protein interaction networks-
dc.typeArticle-
dc.description.naturepublished_or_final_version-
dc.identifier.doi10.1186/1471-2105-15-S2-S9-
dc.identifier.pmid24564855-
dc.identifier.pmcidPMC4015526-
dc.identifier.scopuseid_2-s2.0-84901268258-
dc.identifier.volume15-
dc.identifier.issuesuppl. 2-
dc.identifier.spagearticle no. S9-
dc.identifier.epagearticle no. S9-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats