Predicting binding sites of hydrolase-inhibitor complexes by combining several methods

Taner Z. Sen, Andrzej Kloczkowski, Robert L. Jernigan, Changhui Yan, Vasant Honavar, Kai Ming Ho, Cai Zhuang Wang, Yungok Ihm, Haibo Cao, Xun Gu, Drena Dobbs

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

Background: Protein-protein interactions play a critical role in protein function. Completion of many genomes is being followed rapidly by major efforts to identify interacting protein pairs experimentally in order to decipher the networks of interacting, coordinated-in-action proteins. Identification of protein-protein interaction sites and detection of specific amino acids that contribute to the specificity and the strength of protein interactions is an important problem with broad applications ranging from rational drug design to the analysis of metabolic and signal transduction networks. Results: In order to increase the power of predictive methods for protein-protein interaction sites, we have developed a consensus methodology for combining four different methods. These approaches include: data mining using Support Vector Machines, threading through protein structures, prediction of conserved residues on the protein surface by analysis of phylogenetic trees, and the Conservatism of Conservatism method of Mirny and Shakhnovich. Results obtained on a dataset of hydrolase-inhibitor complexes demonstrate that the combination of all four methods yield improved predictions over the individual methods. Conclusions: We developed a consensus method for predicting protein-protein interface residues by combining sequence and structure-based methods. The success of our consensus approach suggests that similar methodologies can be developed to improve prediction accuracies for other bioinformatic problems.

Original languageEnglish (US)
Article number205
JournalBMC bioinformatics
Volume5
DOIs
StatePublished - Dec 17 2004

Fingerprint

Hydrolases
Binding sites
Inhibitor
Binding Sites
Proteins
Protein
Protein-protein Interaction
Protein Structure Prediction
Drug Design
Signal Transduction
Methodology
Prediction
Phylogenetic Tree
Politics
Specificity
Completion
Amino Acids
Bioinformatics
Support Vector Machine
Data Mining

All Science Journal Classification (ASJC) codes

  • Structural Biology
  • Biochemistry
  • Molecular Biology
  • Computer Science Applications
  • Applied Mathematics

Cite this

Sen, Taner Z. ; Kloczkowski, Andrzej ; Jernigan, Robert L. ; Yan, Changhui ; Honavar, Vasant ; Ho, Kai Ming ; Wang, Cai Zhuang ; Ihm, Yungok ; Cao, Haibo ; Gu, Xun ; Dobbs, Drena. / Predicting binding sites of hydrolase-inhibitor complexes by combining several methods. In: BMC bioinformatics. 2004 ; Vol. 5.
@article{e0314cf962e44ce08907f80bde0dc44e,
title = "Predicting binding sites of hydrolase-inhibitor complexes by combining several methods",
abstract = "Background: Protein-protein interactions play a critical role in protein function. Completion of many genomes is being followed rapidly by major efforts to identify interacting protein pairs experimentally in order to decipher the networks of interacting, coordinated-in-action proteins. Identification of protein-protein interaction sites and detection of specific amino acids that contribute to the specificity and the strength of protein interactions is an important problem with broad applications ranging from rational drug design to the analysis of metabolic and signal transduction networks. Results: In order to increase the power of predictive methods for protein-protein interaction sites, we have developed a consensus methodology for combining four different methods. These approaches include: data mining using Support Vector Machines, threading through protein structures, prediction of conserved residues on the protein surface by analysis of phylogenetic trees, and the Conservatism of Conservatism method of Mirny and Shakhnovich. Results obtained on a dataset of hydrolase-inhibitor complexes demonstrate that the combination of all four methods yield improved predictions over the individual methods. Conclusions: We developed a consensus method for predicting protein-protein interface residues by combining sequence and structure-based methods. The success of our consensus approach suggests that similar methodologies can be developed to improve prediction accuracies for other bioinformatic problems.",
author = "Sen, {Taner Z.} and Andrzej Kloczkowski and Jernigan, {Robert L.} and Changhui Yan and Vasant Honavar and Ho, {Kai Ming} and Wang, {Cai Zhuang} and Yungok Ihm and Haibo Cao and Xun Gu and Drena Dobbs",
year = "2004",
month = "12",
day = "17",
doi = "10.1186/1471-2105-5-205",
language = "English (US)",
volume = "5",
journal = "BMC Bioinformatics",
issn = "1471-2105",
publisher = "BioMed Central",

}

Sen, TZ, Kloczkowski, A, Jernigan, RL, Yan, C, Honavar, V, Ho, KM, Wang, CZ, Ihm, Y, Cao, H, Gu, X & Dobbs, D 2004, 'Predicting binding sites of hydrolase-inhibitor complexes by combining several methods', BMC bioinformatics, vol. 5, 205. https://doi.org/10.1186/1471-2105-5-205

Predicting binding sites of hydrolase-inhibitor complexes by combining several methods. / Sen, Taner Z.; Kloczkowski, Andrzej; Jernigan, Robert L.; Yan, Changhui; Honavar, Vasant; Ho, Kai Ming; Wang, Cai Zhuang; Ihm, Yungok; Cao, Haibo; Gu, Xun; Dobbs, Drena.

In: BMC bioinformatics, Vol. 5, 205, 17.12.2004.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Predicting binding sites of hydrolase-inhibitor complexes by combining several methods

AU - Sen, Taner Z.

AU - Kloczkowski, Andrzej

AU - Jernigan, Robert L.

AU - Yan, Changhui

AU - Honavar, Vasant

AU - Ho, Kai Ming

AU - Wang, Cai Zhuang

AU - Ihm, Yungok

AU - Cao, Haibo

AU - Gu, Xun

AU - Dobbs, Drena

PY - 2004/12/17

Y1 - 2004/12/17

N2 - Background: Protein-protein interactions play a critical role in protein function. Completion of many genomes is being followed rapidly by major efforts to identify interacting protein pairs experimentally in order to decipher the networks of interacting, coordinated-in-action proteins. Identification of protein-protein interaction sites and detection of specific amino acids that contribute to the specificity and the strength of protein interactions is an important problem with broad applications ranging from rational drug design to the analysis of metabolic and signal transduction networks. Results: In order to increase the power of predictive methods for protein-protein interaction sites, we have developed a consensus methodology for combining four different methods. These approaches include: data mining using Support Vector Machines, threading through protein structures, prediction of conserved residues on the protein surface by analysis of phylogenetic trees, and the Conservatism of Conservatism method of Mirny and Shakhnovich. Results obtained on a dataset of hydrolase-inhibitor complexes demonstrate that the combination of all four methods yield improved predictions over the individual methods. Conclusions: We developed a consensus method for predicting protein-protein interface residues by combining sequence and structure-based methods. The success of our consensus approach suggests that similar methodologies can be developed to improve prediction accuracies for other bioinformatic problems.

AB - Background: Protein-protein interactions play a critical role in protein function. Completion of many genomes is being followed rapidly by major efforts to identify interacting protein pairs experimentally in order to decipher the networks of interacting, coordinated-in-action proteins. Identification of protein-protein interaction sites and detection of specific amino acids that contribute to the specificity and the strength of protein interactions is an important problem with broad applications ranging from rational drug design to the analysis of metabolic and signal transduction networks. Results: In order to increase the power of predictive methods for protein-protein interaction sites, we have developed a consensus methodology for combining four different methods. These approaches include: data mining using Support Vector Machines, threading through protein structures, prediction of conserved residues on the protein surface by analysis of phylogenetic trees, and the Conservatism of Conservatism method of Mirny and Shakhnovich. Results obtained on a dataset of hydrolase-inhibitor complexes demonstrate that the combination of all four methods yield improved predictions over the individual methods. Conclusions: We developed a consensus method for predicting protein-protein interface residues by combining sequence and structure-based methods. The success of our consensus approach suggests that similar methodologies can be developed to improve prediction accuracies for other bioinformatic problems.

UR - http://www.scopus.com/inward/record.url?scp=19944430515&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=19944430515&partnerID=8YFLogxK

U2 - 10.1186/1471-2105-5-205

DO - 10.1186/1471-2105-5-205

M3 - Article

C2 - 15606919

AN - SCOPUS:19944430515

VL - 5

JO - BMC Bioinformatics

JF - BMC Bioinformatics

SN - 1471-2105

M1 - 205

ER -