Exploring alternative knowledge representations for protein secondary-structure prediction

Authors:
Uros Midic;A. Keith Dunker;Zoran Obradovic
Affiliations:
Center for Information Science and Technology, Temple University, 1805 N. Broad St., 303 Wachman Hall, Philadelphia, PA 19129, USA.;Center for Computational Biology and Bioinformatics, Indiana University School of Medicine, 714 North Senate Avenue, Suite 250, Indianapolis, IN 46202, USA.;Center for Information Science and Technology, Temple University, 1805 N. Broad St., 303 Wachman Hall, Philadelphia, PA 19129, USA
Venue:
International Journal of Data Mining and Bioinformatics
Year:
2007

Citing 2
Cited 1

Three-stage prediction of protein β-sheets by neural networks, alignments and graph algorithms

Bioinformatics
Bayesian network multi-classifiers for protein secondary structure prediction

Artificial Intelligence in Medicine

A sampling approach for protein backbone fragment conformations

International Journal of Data Mining and Bioinformatics

Quantified Score

Hi-index	0.00

Visualization

Abstract

Methods for 3-class secondary-structure prediction are thoughtto be reaching the highest achievable accuracy. Their accuracy onβ-sheet residue class is considerably lower than for the othertwo classes. We analysed the relevance of 315 individual inputattributes for a predictor with the usual framework of usingsequence-profile based data with an input window of fixed size. Wepropose two alternative knowledge representations withsignificantly smaller sets of input attributes. We alsoinvestigated the possibility of exploiting the prediction ofconnected pairs of β-sheet residues and the prediction ofresidue contact maps for the improvement of accuracy ofsecondary-structure prediction.