CSSP2: An improved method for predicting contact-dependent secondary structure propensity

Sukjoon Yoon, William J. Welsh, Heeyoung Jung, Young Do Yoo

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

The calculation of contact-dependent secondary structure propensity (CSSP) has been reported to sensitively detect non-native β-strand propensities in the core sequences of amyloidogenic proteins. Here we describe a noble energy-based CSSP method implemented on dual artificial neural networks that rapidly and accurately estimate the potential for the non-native secondary structure formation in local regions of protein sequences. In this method, we attempted to quantify long-range interaction patterns in diverse secondary structures by potential energy calculations and decomposition on a pairwise per-residue basis. The calculated energy parameters and seven-residue sequence information were used as inputs for artificial neural networks (ANNs) to predict sequence potential for secondary structure conversion. The trained single ANN using the >(i, i ± 4) interaction energy parameter exhibited 74% accuracy in predicting the secondary structure of test sequences in their native energy state, while the dual ANN-based predictor using (i, i ± 4) and >(i, i ± 4) interaction energies showed 83% prediction accuracy. The present method provides a simple and accurate tool for predicting sequence potential for secondary structure conversions without using 3D structural information.

Original languageEnglish
Pages (from-to)373-377
Number of pages5
JournalComputational Biology and Chemistry
Volume31
Issue number5-6
DOIs
Publication statusPublished - 2007 Oct 1

Fingerprint

Secondary Structure
Contact
Neural networks
Dependent
Artificial Neural Network
Amyloidogenic Proteins
Energy
energy
proteins
Potential energy
Electron energy levels
interactions
predictions
strands
Decomposition
Proteins
Long-range Interactions
Protein Sequence
potential energy
Interaction

Keywords

  • Amyloid fibril formation
  • Artificial neural network
  • Energy decomposition
  • Machine learning
  • Secondary structure prediction

ASJC Scopus subject areas

  • Biochemistry
  • Structural Biology
  • Analytical Chemistry
  • Physical and Theoretical Chemistry

Cite this

CSSP2 : An improved method for predicting contact-dependent secondary structure propensity. / Yoon, Sukjoon; Welsh, William J.; Jung, Heeyoung; Yoo, Young Do.

In: Computational Biology and Chemistry, Vol. 31, No. 5-6, 01.10.2007, p. 373-377.

Research output: Contribution to journalArticle

Yoon, Sukjoon ; Welsh, William J. ; Jung, Heeyoung ; Yoo, Young Do. / CSSP2 : An improved method for predicting contact-dependent secondary structure propensity. In: Computational Biology and Chemistry. 2007 ; Vol. 31, No. 5-6. pp. 373-377.
@article{db67230c5626411ea242afd1047b6b70,
title = "CSSP2: An improved method for predicting contact-dependent secondary structure propensity",
abstract = "The calculation of contact-dependent secondary structure propensity (CSSP) has been reported to sensitively detect non-native β-strand propensities in the core sequences of amyloidogenic proteins. Here we describe a noble energy-based CSSP method implemented on dual artificial neural networks that rapidly and accurately estimate the potential for the non-native secondary structure formation in local regions of protein sequences. In this method, we attempted to quantify long-range interaction patterns in diverse secondary structures by potential energy calculations and decomposition on a pairwise per-residue basis. The calculated energy parameters and seven-residue sequence information were used as inputs for artificial neural networks (ANNs) to predict sequence potential for secondary structure conversion. The trained single ANN using the >(i, i ± 4) interaction energy parameter exhibited 74{\%} accuracy in predicting the secondary structure of test sequences in their native energy state, while the dual ANN-based predictor using (i, i ± 4) and >(i, i ± 4) interaction energies showed 83{\%} prediction accuracy. The present method provides a simple and accurate tool for predicting sequence potential for secondary structure conversions without using 3D structural information.",
keywords = "Amyloid fibril formation, Artificial neural network, Energy decomposition, Machine learning, Secondary structure prediction",
author = "Sukjoon Yoon and Welsh, {William J.} and Heeyoung Jung and Yoo, {Young Do}",
year = "2007",
month = "10",
day = "1",
doi = "10.1016/j.compbiolchem.2007.06.002",
language = "English",
volume = "31",
pages = "373--377",
journal = "Computational Biology and Chemistry",
issn = "1476-9271",
publisher = "Elsevier Limited",
number = "5-6",

}

TY - JOUR

T1 - CSSP2

T2 - An improved method for predicting contact-dependent secondary structure propensity

AU - Yoon, Sukjoon

AU - Welsh, William J.

AU - Jung, Heeyoung

AU - Yoo, Young Do

PY - 2007/10/1

Y1 - 2007/10/1

N2 - The calculation of contact-dependent secondary structure propensity (CSSP) has been reported to sensitively detect non-native β-strand propensities in the core sequences of amyloidogenic proteins. Here we describe a noble energy-based CSSP method implemented on dual artificial neural networks that rapidly and accurately estimate the potential for the non-native secondary structure formation in local regions of protein sequences. In this method, we attempted to quantify long-range interaction patterns in diverse secondary structures by potential energy calculations and decomposition on a pairwise per-residue basis. The calculated energy parameters and seven-residue sequence information were used as inputs for artificial neural networks (ANNs) to predict sequence potential for secondary structure conversion. The trained single ANN using the >(i, i ± 4) interaction energy parameter exhibited 74% accuracy in predicting the secondary structure of test sequences in their native energy state, while the dual ANN-based predictor using (i, i ± 4) and >(i, i ± 4) interaction energies showed 83% prediction accuracy. The present method provides a simple and accurate tool for predicting sequence potential for secondary structure conversions without using 3D structural information.

AB - The calculation of contact-dependent secondary structure propensity (CSSP) has been reported to sensitively detect non-native β-strand propensities in the core sequences of amyloidogenic proteins. Here we describe a noble energy-based CSSP method implemented on dual artificial neural networks that rapidly and accurately estimate the potential for the non-native secondary structure formation in local regions of protein sequences. In this method, we attempted to quantify long-range interaction patterns in diverse secondary structures by potential energy calculations and decomposition on a pairwise per-residue basis. The calculated energy parameters and seven-residue sequence information were used as inputs for artificial neural networks (ANNs) to predict sequence potential for secondary structure conversion. The trained single ANN using the >(i, i ± 4) interaction energy parameter exhibited 74% accuracy in predicting the secondary structure of test sequences in their native energy state, while the dual ANN-based predictor using (i, i ± 4) and >(i, i ± 4) interaction energies showed 83% prediction accuracy. The present method provides a simple and accurate tool for predicting sequence potential for secondary structure conversions without using 3D structural information.

KW - Amyloid fibril formation

KW - Artificial neural network

KW - Energy decomposition

KW - Machine learning

KW - Secondary structure prediction

UR - http://www.scopus.com/inward/record.url?scp=35148821507&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=35148821507&partnerID=8YFLogxK

U2 - 10.1016/j.compbiolchem.2007.06.002

DO - 10.1016/j.compbiolchem.2007.06.002

M3 - Article

C2 - 17644485

AN - SCOPUS:35148821507

VL - 31

SP - 373

EP - 377

JO - Computational Biology and Chemistry

JF - Computational Biology and Chemistry

SN - 1476-9271

IS - 5-6

ER -