Neural Feature Association Rule Mining for Protein Interaction Prediction
The prediction of protein interactions is an important problem in post–genomic biology. In this paper, we present an association rule mining method for protein interaction prediction. A neural network is used to cluster protein interaction data and a feature selection is used to reduce the dimension of protein features. For model training, the preliminary network model was constructed with existing protein interaction data in terms of their functional categories and interactions. A set of association rules for protein interaction prediction are derived by decoding a set of learned weights of trained neural network after this model training. The protein interaction data of Yeast from public databases are used. The prediction performance was compared with simple association rule-based approach. According to the experimental results, proposed method achieved about 95.5% accuracy.
KeywordsAssociation Rule Association Rule Mining Trained Neural Network Adaptive Resonance Theory Adaptive Neural Network
Unable to display preview. Download preview PDF.
- 6.Barbara, M.: ART1 and pattern clustering. In: Proceedings of the Summer School of Connectionist Models, pp. 174–185. Morgan Kaufmann, San Francisco (1988)Google Scholar
- 7.Heins, L.G., Tauritz, D.R.: Adaptive Resonance Theory (ART): An Introduction. Internal Report 95-35, Dept. of Computer Science, Leiden University, Netherlands, 174–85 (1995)Google Scholar
- 8.Rangarajan, et al.: Adaptive Neural Network Clustering of Web Users. IEEE Computer 37(4), 34–40 (2004)Google Scholar
- 9.Yu, L., Liu, H.: Feature Selection for High Dimensional Data: A Fast Correlation-based Filter Solution. In: Proceeding of ICML 2003, pp. 856–863 (2003)Google Scholar
- 13.Eom, J.-H.: Prediction of Yeast Protein–Protein Interactions by Neural Association Rule. Internal Report 04-03, Dept. of Computer Sci.&Eng., Seoul National University, Republic of Korea, 1–12 (2004)Google Scholar