Introducing novel physicochemical based features to enhance protein fold prediction accuracy

Abdollah Dehzangi, Bahador Ganjeh Khosravi

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

3 Scopus citations

Abstract

One of the most important goals of bioinformatics is the ability to predict tertiary structure of a protein from its amino acids sequence. In pattern recognition terms, protein fold prediction problem can be mapped to a multi class classification problem. Feature analysis and pattern classification techniques have been applied on proteins' sequence to reveal underlying tertiary structures. In this paper, new features are proposed based on the physicochemical properties of the amino acids. Proposed features are extracted by using a modified feature extraction method adopted by Dubchak et al. AdaBoost.M1 and Multi Layer Perceptron attained good result constantly with existing features. Therefore, we assess the prediction accuracy of our new features and modified feature extraction with these methods. Our experimental results showed that the new features and modified feature extraction method outperformed the prediction accuracy that was reported in previous works.

Original languageEnglish (US)
Title of host publication2010 International Conference on Computer Design and Applications, ICCDA 2010
PagesV1592-V1596
DOIs
StatePublished - 2010
Externally publishedYes
Event2010 International Conference on Computer Design and Applications, ICCDA 2010 - Qinhuangdao, Hebei, China
Duration: Jun 25 2010Jun 27 2010

Publication series

Name2010 International Conference on Computer Design and Applications, ICCDA 2010
Volume1

Conference

Conference2010 International Conference on Computer Design and Applications, ICCDA 2010
Country/TerritoryChina
CityQinhuangdao, Hebei
Period6/25/106/27/10

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Science Applications

Keywords

  • AdaBoost.M1
  • Distribution factor
  • Feature extraction
  • Multi layer perceptron
  • Protein fold prediction problem
  • Secondary structure
  • Size of amino acids
  • Turn
  • α -helix
  • β -strand

Fingerprint

Dive into the research topics of 'Introducing novel physicochemical based features to enhance protein fold prediction accuracy'. Together they form a unique fingerprint.

Cite this