Predicting protein secondary structure with a nearest-neighbor algorithm

Steven L Salzberg, Scott Cost

Research output: Contribution to journalArticle

Abstract

We have developed a new method for protein secondary structure prediction that achieves accuracies as high as 71.0%, the highest value yet reported. The main component of our method is a nearest-neighbor algorithm that uses a more sophisticated treatment of the feature space than standard nearest-neighbor methods. It calculates distance tables that allow it to produce real-valued distances between amino acid residues, and attaches weights to the instances to further modify the the structure of feature space. The algorithm, which is closely related to the memory-based reasoning method of Zhang et al., is simple and easy to train, and has also been applied with excellent results to the problem of identifying DNA promoter sequences.

Original languageEnglish (US)
Pages (from-to)371-374
Number of pages4
JournalJournal of Molecular Biology
Volume227
Issue number2
DOIs
StatePublished - Sep 20 1992

Fingerprint

Secondary Protein Structure
Amino Acids
Weights and Measures

Keywords

  • memory-based reasoning
  • nearest-neighbor methods
  • neural nets
  • protein secondary structure

ASJC Scopus subject areas

  • Virology

Cite this

Predicting protein secondary structure with a nearest-neighbor algorithm. / Salzberg, Steven L; Cost, Scott.

In: Journal of Molecular Biology, Vol. 227, No. 2, 20.09.1992, p. 371-374.

Research output: Contribution to journalArticle

@article{e3be13aa90314879a695efa637765b18,
title = "Predicting protein secondary structure with a nearest-neighbor algorithm",
abstract = "We have developed a new method for protein secondary structure prediction that achieves accuracies as high as 71.0{\%}, the highest value yet reported. The main component of our method is a nearest-neighbor algorithm that uses a more sophisticated treatment of the feature space than standard nearest-neighbor methods. It calculates distance tables that allow it to produce real-valued distances between amino acid residues, and attaches weights to the instances to further modify the the structure of feature space. The algorithm, which is closely related to the memory-based reasoning method of Zhang et al., is simple and easy to train, and has also been applied with excellent results to the problem of identifying DNA promoter sequences.",
keywords = "memory-based reasoning, nearest-neighbor methods, neural nets, protein secondary structure",
author = "Salzberg, {Steven L} and Scott Cost",
year = "1992",
month = "9",
day = "20",
doi = "10.1016/0022-2836(92)90892-N",
language = "English (US)",
volume = "227",
pages = "371--374",
journal = "Journal of Molecular Biology",
issn = "0022-2836",
publisher = "Academic Press Inc.",
number = "2",

}

TY - JOUR

T1 - Predicting protein secondary structure with a nearest-neighbor algorithm

AU - Salzberg, Steven L

AU - Cost, Scott

PY - 1992/9/20

Y1 - 1992/9/20

N2 - We have developed a new method for protein secondary structure prediction that achieves accuracies as high as 71.0%, the highest value yet reported. The main component of our method is a nearest-neighbor algorithm that uses a more sophisticated treatment of the feature space than standard nearest-neighbor methods. It calculates distance tables that allow it to produce real-valued distances between amino acid residues, and attaches weights to the instances to further modify the the structure of feature space. The algorithm, which is closely related to the memory-based reasoning method of Zhang et al., is simple and easy to train, and has also been applied with excellent results to the problem of identifying DNA promoter sequences.

AB - We have developed a new method for protein secondary structure prediction that achieves accuracies as high as 71.0%, the highest value yet reported. The main component of our method is a nearest-neighbor algorithm that uses a more sophisticated treatment of the feature space than standard nearest-neighbor methods. It calculates distance tables that allow it to produce real-valued distances between amino acid residues, and attaches weights to the instances to further modify the the structure of feature space. The algorithm, which is closely related to the memory-based reasoning method of Zhang et al., is simple and easy to train, and has also been applied with excellent results to the problem of identifying DNA promoter sequences.

KW - memory-based reasoning

KW - nearest-neighbor methods

KW - neural nets

KW - protein secondary structure

UR - http://www.scopus.com/inward/record.url?scp=0026674246&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0026674246&partnerID=8YFLogxK

U2 - 10.1016/0022-2836(92)90892-N

DO - 10.1016/0022-2836(92)90892-N

M3 - Article

C2 - 1404357

AN - SCOPUS:0026674246

VL - 227

SP - 371

EP - 374

JO - Journal of Molecular Biology

JF - Journal of Molecular Biology

SN - 0022-2836

IS - 2

ER -