Protein secondary structure prediction based on an improved support vector machines approach

Hyunsoo Kim; Haesun Park

doi:10.1093/protein/gzg072

Protein secondary structure prediction based on an improved support vector machines approach

Hyunsoo Kim, Haesun Park

Research output: Contribution to journal › Article › peer-review

177 Scopus citations

Abstract

The prediction of protein secondary structure is an important step in the prediction of protein tertiary structure. A new protein secondary structure prediction method, SVMpsi, was developed to improve the current level of prediction by incorporating new tertiary classifiers and their jury decision system, and the PSI-BLAST PSSM profiles. Additionally, efficient methods to handle unbalanced data and a new optimization strategy for maximizing the Q ₃ measure were developed. The SVMpsi produces the highest published Q₃ and SOV94 scores on both the RS126 and CB513 data sets to date. For a new KP480 set, the prediction accuracy of SVMpsi was Q₃ = 78. 5% and SOV94 = 82.8%. Moreover, the blind test results for 136 non-redundant protein sequences which do not contain homologues of training data sets were Q₃ = 77.2% and SOV94 = 81.8%. The SVMpsi results in CASP5 illustrate that it is another competitive method to predict protein secondary structure.

Original language	English (US)
Pages (from-to)	553-560
Number of pages	8
Journal	Protein Engineering
Volume	16
Issue number	8
DOIs	https://doi.org/10.1093/protein/gzg072
State	Published - Aug 1 2003
Externally published	Yes

Bibliographical note

Funding Information:
This work was supported in part by the National Science Foundation, grants CCR-0204109 and ACI-0305543. Any opinions, findings and conclusions or recommendations expressed in this paper are those of the authors and do not necessarily reflect the views of the National Science Foundation. Part of this work was carried out while the authors were visiting the Korea Institute for Advanced Study, Seoul, Korea, from January 2002 to August 2002. The authors thank the University of Minnesota Supercomputing Institute (MSI) for intensive numerical computing. They also thank Professor Thorsten Joachims for making SVMlight software available, James A.Cuff and Professor Geoffrey J.Barton for providing the data set and Professor David T.Jones for the PFILT software and his kind help.

Keywords

Directed acyclic graph scheme
Position-specific scoring matrix
Protein structure prediction
Secondary structure
Support vector machines

Publisher link

10.1093/protein/gzg072

Find It

Find It at U of M Libraries

Cite this

@article{fb9f867470064313b863a88c8cc673ba,

title = "Protein secondary structure prediction based on an improved support vector machines approach",

abstract = "The prediction of protein secondary structure is an important step in the prediction of protein tertiary structure. A new protein secondary structure prediction method, SVMpsi, was developed to improve the current level of prediction by incorporating new tertiary classifiers and their jury decision system, and the PSI-BLAST PSSM profiles. Additionally, efficient methods to handle unbalanced data and a new optimization strategy for maximizing the Q 3 measure were developed. The SVMpsi produces the highest published Q3 and SOV94 scores on both the RS126 and CB513 data sets to date. For a new KP480 set, the prediction accuracy of SVMpsi was Q3 = 78. 5% and SOV94 = 82.8%. Moreover, the blind test results for 136 non-redundant protein sequences which do not contain homologues of training data sets were Q3 = 77.2% and SOV94 = 81.8%. The SVMpsi results in CASP5 illustrate that it is another competitive method to predict protein secondary structure.",

keywords = "Directed acyclic graph scheme, Position-specific scoring matrix, Protein structure prediction, Secondary structure, Support vector machines",

author = "Hyunsoo Kim and Haesun Park",

note = "Funding Information: This work was supported in part by the National Science Foundation, grants CCR-0204109 and ACI-0305543. Any opinions, findings and conclusions or recommendations expressed in this paper are those of the authors and do not necessarily reflect the views of the National Science Foundation. Part of this work was carried out while the authors were visiting the Korea Institute for Advanced Study, Seoul, Korea, from January 2002 to August 2002. The authors thank the University of Minnesota Supercomputing Institute (MSI) for intensive numerical computing. They also thank Professor Thorsten Joachims for making SVMlight software available, James A.Cuff and Professor Geoffrey J.Barton for providing the data set and Professor David T.Jones for the PFILT software and his kind help.",

year = "2003",

month = aug,

day = "1",

doi = "10.1093/protein/gzg072",

language = "English (US)",

volume = "16",

pages = "553--560",

journal = "Protein Engineering",

issn = "0269-2139",

publisher = "Oxford University Press",

number = "8",

}

TY - JOUR

T1 - Protein secondary structure prediction based on an improved support vector machines approach

AU - Kim, Hyunsoo

AU - Park, Haesun

N1 - Funding Information: This work was supported in part by the National Science Foundation, grants CCR-0204109 and ACI-0305543. Any opinions, findings and conclusions or recommendations expressed in this paper are those of the authors and do not necessarily reflect the views of the National Science Foundation. Part of this work was carried out while the authors were visiting the Korea Institute for Advanced Study, Seoul, Korea, from January 2002 to August 2002. The authors thank the University of Minnesota Supercomputing Institute (MSI) for intensive numerical computing. They also thank Professor Thorsten Joachims for making SVMlight software available, James A.Cuff and Professor Geoffrey J.Barton for providing the data set and Professor David T.Jones for the PFILT software and his kind help.

PY - 2003/8/1

Y1 - 2003/8/1

N2 - The prediction of protein secondary structure is an important step in the prediction of protein tertiary structure. A new protein secondary structure prediction method, SVMpsi, was developed to improve the current level of prediction by incorporating new tertiary classifiers and their jury decision system, and the PSI-BLAST PSSM profiles. Additionally, efficient methods to handle unbalanced data and a new optimization strategy for maximizing the Q 3 measure were developed. The SVMpsi produces the highest published Q3 and SOV94 scores on both the RS126 and CB513 data sets to date. For a new KP480 set, the prediction accuracy of SVMpsi was Q3 = 78. 5% and SOV94 = 82.8%. Moreover, the blind test results for 136 non-redundant protein sequences which do not contain homologues of training data sets were Q3 = 77.2% and SOV94 = 81.8%. The SVMpsi results in CASP5 illustrate that it is another competitive method to predict protein secondary structure.

AB - The prediction of protein secondary structure is an important step in the prediction of protein tertiary structure. A new protein secondary structure prediction method, SVMpsi, was developed to improve the current level of prediction by incorporating new tertiary classifiers and their jury decision system, and the PSI-BLAST PSSM profiles. Additionally, efficient methods to handle unbalanced data and a new optimization strategy for maximizing the Q 3 measure were developed. The SVMpsi produces the highest published Q3 and SOV94 scores on both the RS126 and CB513 data sets to date. For a new KP480 set, the prediction accuracy of SVMpsi was Q3 = 78. 5% and SOV94 = 82.8%. Moreover, the blind test results for 136 non-redundant protein sequences which do not contain homologues of training data sets were Q3 = 77.2% and SOV94 = 81.8%. The SVMpsi results in CASP5 illustrate that it is another competitive method to predict protein secondary structure.

KW - Directed acyclic graph scheme

KW - Position-specific scoring matrix

KW - Protein structure prediction

KW - Secondary structure

KW - Support vector machines

UR - http://www.scopus.com/inward/record.url?scp=0141593924&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0141593924&partnerID=8YFLogxK

U2 - 10.1093/protein/gzg072

DO - 10.1093/protein/gzg072

M3 - Article

C2 - 12968073

AN - SCOPUS:0141593924

SN - 0269-2139

VL - 16

SP - 553

EP - 560

JO - Protein Engineering

JF - Protein Engineering

IS - 8

ER -

Protein secondary structure prediction based on an improved support vector machines approach

Abstract

Bibliographical note

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this