Kernel methods for predicting protein--protein interactions
Bioinformatics
Domain-based predictive models for protein-protein interaction prediction
EURASIP Journal on Applied Signal Processing
IWANN '09 Proceedings of the 10th International Work-Conference on Artificial Neural Networks: Part I: Bio-Inspired Systems: Computational and Ambient Intelligence
Computers in Biology and Medicine
Computers in Biology and Medicine
Information Sciences: an International Journal
Research Article: Bioinformatic analysis of molecular network of glucosinolate biosynthesis
Computational Biology and Chemistry
LIBSVM: A library for support vector machines
ACM Transactions on Intelligent Systems and Technology (TIST)
Searching a pattern in compressed DNA sequences
International Journal of Bioinformatics Research and Applications
Using support vector machine for characteristics prediction of hydraulic valve
International Journal of Computer Applications in Technology
Application with a hybrid ant colony optimisation in motif detecting problem
International Journal of Computer Applications in Technology
An ant colony optimisation algorithm for constructing phylogenetic tree
International Journal of Computer Applications in Technology
Hi-index | 0.00 |
Protein-protein interactions PPIs are of biological interest because they orchestrate a number of cellular processes such as metabolic pathways and immunological recognition. This paper aims at exploring more and removing PPIs falsely predicted PPIs involved in glucosinolate biosynthesis in Arabidopsis. A symmetric kernel function is proposed according to the approach of feature representation which combines the domain and domain-domain interaction DDI information in this paper. The performance of this kernel indicates SVM based PPIs predictor trained with this kernel is highly effective. According to the prediction result, proteins with Arabidopsis Genome Initiative AGI numbers AT4G14800 and AT5G54810, AT5G05730 and AT4G18040, AT1G04510 and AT5G05260 are affirmed as interactive among the 237 low level of confidence PPIs pairs. Furthermore, the SVM-based PPIs predictor is used to explore PPIs of AT1G74090 and AT5G07690 both of which are members of the four glucosinolate biosynthesis pathway proteins absent from AtPIN.