Using a stochastic adaboost algorithm to discover interactome motif pairs from sequences

Authors:
Huan Yu;Minping Qian;Minghua Deng
Affiliations:
LMAM, School of Mathematical Sciences and Center for Theoretical Biology, Peking University, Beijing, P.R. China;LMAM, School of Mathematical Sciences and Center for Theoretical Biology, Peking University, Beijing, P.R. China;LMAM, School of Mathematical Sciences and Center for Theoretical Biology, Peking University, Beijing, P.R. China
Venue:
ICIC'06 Proceedings of the 2006 international conference on Computational Intelligence and Bioinformatics - Volume Part III
Year:
2006

Citing 7
Cited 0

A decision-theoretic generalization of on-line learning and an application to boosting

Journal of Computer and System Sciences - Special issue: 26th annual ACM symposium on the theory of computing & STOC'94, May 23–25, 1994, and second annual Europe an conference on computational learning theory (EuroCOLT'95), March 13–15, 1995
Learning Quickly When Irrelevant Attributes Abound: A New Linear-Threshold Algorithm

Machine Learning
Statistical analysis of domains in interacting protein pairs

Bioinformatics
Inferring protein–protein interactions through high-throughput interaction data from diverse organisms

Bioinformatics
Prediction of protein–protein interactions using distant conservation of sequence patterns and structure relationships

Bioinformatics
Prediction of protein--protein interactions using random decision forest framework

Bioinformatics
A regularized discriminative model for the prediction of protein--peptide interactions

Bioinformatics

Quantified Score

Hi-index	0.00

Visualization

Abstract

Protein interactome is an important research focus in the post-genomic era. The identification of interacting motif pairs is essential for exploring the mechanism of protein interactions. We describe a stochastic AdaBoost approach for discovering motif pairs from known interactions and pairs of proteins that are putatively not to interact. Our interacting motif pairs are validated by multiple-chain PDB structures and show more significant than those selected by traditional statistical method. Furthermore, in a cross-validated comparison, our model can be used to predict interactions between proteins with higher sensitivity (66.42%) and specificity (87.38%) comparing with the Naive Bayes model and the dominating model.