Comparing kernels for predicting protein binding sites from amino acid sequence

Feihong Wu, Byron Olson, Drena Dobbs, Vasant Honavar

Research output: Chapter in Book/Report/Conference proceedingConference contribution

7 Citations (Scopus)

Abstract

The ability to identify protein binding sites and to detect specific amino acid residues that contribute to the specificity and affinity of protein interactions has important implications for problems ranging from rational drug design to analysis of metabolic and signal transduction networks. Support vector machines (SVM) and related kernel methods offer an attractive approach to predicting protein binding sites. An appropriate choice of the kernel function is critical to the performance of SVM. Kernel functions offer a way to incorporate domain-specific knowledge into the classifier. We compare the performance of three types of kernels functions: identity kernel, sequence-alignment kernel, and amino acid substitution matrix kernel in the case of SVM classifiers for predicting protein-protein, protein-DNA and protein-RNA binding sites. The results show that the identity kernel is quite effective in on all three tasks. The substitution kernel based on amino acid substitution matrices that take into account structural or evolutionary conservation or physicochemical properties of amino acids yields modest improvement.

Original languageEnglish (US)
Title of host publicationInternational Joint Conference on Neural Networks 2006, IJCNN '06
Pages1612-1616
Number of pages5
StatePublished - 2006
EventInternational Joint Conference on Neural Networks 2006, IJCNN '06 - Vancouver, BC, Canada
Duration: Jul 16 2006Jul 21 2006

Other

OtherInternational Joint Conference on Neural Networks 2006, IJCNN '06
CountryCanada
CityVancouver, BC
Period7/16/067/21/06

Fingerprint

Binding sites
Amino acids
Proteins
Support vector machines
Substitution reactions
Classifiers
Signal transduction
RNA
Conservation
DNA
Protein Binding

All Science Journal Classification (ASJC) codes

  • Software

Cite this

Wu, F., Olson, B., Dobbs, D., & Honavar, V. (2006). Comparing kernels for predicting protein binding sites from amino acid sequence. In International Joint Conference on Neural Networks 2006, IJCNN '06 (pp. 1612-1616). [1716299]
Wu, Feihong ; Olson, Byron ; Dobbs, Drena ; Honavar, Vasant. / Comparing kernels for predicting protein binding sites from amino acid sequence. International Joint Conference on Neural Networks 2006, IJCNN '06. 2006. pp. 1612-1616
@inproceedings{406465092ef541fc8a0e5d953b955b8c,
title = "Comparing kernels for predicting protein binding sites from amino acid sequence",
abstract = "The ability to identify protein binding sites and to detect specific amino acid residues that contribute to the specificity and affinity of protein interactions has important implications for problems ranging from rational drug design to analysis of metabolic and signal transduction networks. Support vector machines (SVM) and related kernel methods offer an attractive approach to predicting protein binding sites. An appropriate choice of the kernel function is critical to the performance of SVM. Kernel functions offer a way to incorporate domain-specific knowledge into the classifier. We compare the performance of three types of kernels functions: identity kernel, sequence-alignment kernel, and amino acid substitution matrix kernel in the case of SVM classifiers for predicting protein-protein, protein-DNA and protein-RNA binding sites. The results show that the identity kernel is quite effective in on all three tasks. The substitution kernel based on amino acid substitution matrices that take into account structural or evolutionary conservation or physicochemical properties of amino acids yields modest improvement.",
author = "Feihong Wu and Byron Olson and Drena Dobbs and Vasant Honavar",
year = "2006",
language = "English (US)",
isbn = "0780394909",
pages = "1612--1616",
booktitle = "International Joint Conference on Neural Networks 2006, IJCNN '06",

}

Wu, F, Olson, B, Dobbs, D & Honavar, V 2006, Comparing kernels for predicting protein binding sites from amino acid sequence. in International Joint Conference on Neural Networks 2006, IJCNN '06., 1716299, pp. 1612-1616, International Joint Conference on Neural Networks 2006, IJCNN '06, Vancouver, BC, Canada, 7/16/06.

Comparing kernels for predicting protein binding sites from amino acid sequence. / Wu, Feihong; Olson, Byron; Dobbs, Drena; Honavar, Vasant.

International Joint Conference on Neural Networks 2006, IJCNN '06. 2006. p. 1612-1616 1716299.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Comparing kernels for predicting protein binding sites from amino acid sequence

AU - Wu, Feihong

AU - Olson, Byron

AU - Dobbs, Drena

AU - Honavar, Vasant

PY - 2006

Y1 - 2006

N2 - The ability to identify protein binding sites and to detect specific amino acid residues that contribute to the specificity and affinity of protein interactions has important implications for problems ranging from rational drug design to analysis of metabolic and signal transduction networks. Support vector machines (SVM) and related kernel methods offer an attractive approach to predicting protein binding sites. An appropriate choice of the kernel function is critical to the performance of SVM. Kernel functions offer a way to incorporate domain-specific knowledge into the classifier. We compare the performance of three types of kernels functions: identity kernel, sequence-alignment kernel, and amino acid substitution matrix kernel in the case of SVM classifiers for predicting protein-protein, protein-DNA and protein-RNA binding sites. The results show that the identity kernel is quite effective in on all three tasks. The substitution kernel based on amino acid substitution matrices that take into account structural or evolutionary conservation or physicochemical properties of amino acids yields modest improvement.

AB - The ability to identify protein binding sites and to detect specific amino acid residues that contribute to the specificity and affinity of protein interactions has important implications for problems ranging from rational drug design to analysis of metabolic and signal transduction networks. Support vector machines (SVM) and related kernel methods offer an attractive approach to predicting protein binding sites. An appropriate choice of the kernel function is critical to the performance of SVM. Kernel functions offer a way to incorporate domain-specific knowledge into the classifier. We compare the performance of three types of kernels functions: identity kernel, sequence-alignment kernel, and amino acid substitution matrix kernel in the case of SVM classifiers for predicting protein-protein, protein-DNA and protein-RNA binding sites. The results show that the identity kernel is quite effective in on all three tasks. The substitution kernel based on amino acid substitution matrices that take into account structural or evolutionary conservation or physicochemical properties of amino acids yields modest improvement.

UR - http://www.scopus.com/inward/record.url?scp=40649099150&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=40649099150&partnerID=8YFLogxK

M3 - Conference contribution

SN - 0780394909

SN - 9780780394902

SP - 1612

EP - 1616

BT - International Joint Conference on Neural Networks 2006, IJCNN '06

ER -

Wu F, Olson B, Dobbs D, Honavar V. Comparing kernels for predicting protein binding sites from amino acid sequence. In International Joint Conference on Neural Networks 2006, IJCNN '06. 2006. p. 1612-1616. 1716299