Structural Bioinformatics Prediction of Membrane-binding Proteins

Nitin Bhardwaj, Robert Stahelin, Robert E. Langlois, Wonhwa Cho, Hui Lu

Research output: Contribution to journalArticle

45 Citations (Scopus)

Abstract

Membrane-binding peripheral proteins play important roles in many biological processes, including cell signaling and membrane trafficking. Unlike integral membrane proteins, these proteins bind the membrane mostly in a reversible manner. Since peripheral proteins do not have canonical transmembrane segments, it is difficult to identify them from their amino acid sequences. As a first step toward genome-scale identification of membrane-binding peripheral proteins, we built a kernel-based machine learning protocol. Key features of known membrane-binding proteins, including electrostatic properties and amino acid composition, were calculated from their amino acid sequences and tertiary structures, which were then incorporated into the support vector machine to perform the classification. A data set of 40 membrane-binding proteins and 230 non-membrane-binding proteins was used to construct and validate the protocol. Cross-validation and holdout evaluation of the protocol showed that the accuracy of the prediction reached up to 93.7% and 91.6%, respectively. The protocol was applied to the prediction of membrane-binding properties of four C2 domains from novel protein kinases C. Although these C2 domains have 50% sequence identity, only one of them was predicted to bind the membrane, which was verified experimentally with surface plasmon resonance analysis. These results suggest that our protocol can be used for predicting membrane-binding properties of a wide variety of modular domains and may be further extended to genome-scale identification of membrane-binding peripheral proteins.

Original languageEnglish (US)
Pages (from-to)486-495
Number of pages10
JournalJournal of Molecular Biology
Volume359
Issue number2
DOIs
StatePublished - Jun 2 2006
Externally publishedYes

Fingerprint

Computational Biology
Carrier Proteins
Membrane Proteins
Membranes
Amino Acid Sequence
Genome
Biological Phenomena
Surface Plasmon Resonance
Static Electricity
Protein Kinase C
Cell Membrane
Amino Acids
Proteins

Keywords

  • function annotation
  • peripheral proteins
  • protein function prediction
  • protein-membrane interactions
  • support vector machines

ASJC Scopus subject areas

  • Virology

Cite this

Structural Bioinformatics Prediction of Membrane-binding Proteins. / Bhardwaj, Nitin; Stahelin, Robert; Langlois, Robert E.; Cho, Wonhwa; Lu, Hui.

In: Journal of Molecular Biology, Vol. 359, No. 2, 02.06.2006, p. 486-495.

Research output: Contribution to journalArticle

Bhardwaj, Nitin ; Stahelin, Robert ; Langlois, Robert E. ; Cho, Wonhwa ; Lu, Hui. / Structural Bioinformatics Prediction of Membrane-binding Proteins. In: Journal of Molecular Biology. 2006 ; Vol. 359, No. 2. pp. 486-495.
@article{963a3d007d2c4ff591732ddaf9dfac9d,
title = "Structural Bioinformatics Prediction of Membrane-binding Proteins",
abstract = "Membrane-binding peripheral proteins play important roles in many biological processes, including cell signaling and membrane trafficking. Unlike integral membrane proteins, these proteins bind the membrane mostly in a reversible manner. Since peripheral proteins do not have canonical transmembrane segments, it is difficult to identify them from their amino acid sequences. As a first step toward genome-scale identification of membrane-binding peripheral proteins, we built a kernel-based machine learning protocol. Key features of known membrane-binding proteins, including electrostatic properties and amino acid composition, were calculated from their amino acid sequences and tertiary structures, which were then incorporated into the support vector machine to perform the classification. A data set of 40 membrane-binding proteins and 230 non-membrane-binding proteins was used to construct and validate the protocol. Cross-validation and holdout evaluation of the protocol showed that the accuracy of the prediction reached up to 93.7{\%} and 91.6{\%}, respectively. The protocol was applied to the prediction of membrane-binding properties of four C2 domains from novel protein kinases C. Although these C2 domains have 50{\%} sequence identity, only one of them was predicted to bind the membrane, which was verified experimentally with surface plasmon resonance analysis. These results suggest that our protocol can be used for predicting membrane-binding properties of a wide variety of modular domains and may be further extended to genome-scale identification of membrane-binding peripheral proteins.",
keywords = "function annotation, peripheral proteins, protein function prediction, protein-membrane interactions, support vector machines",
author = "Nitin Bhardwaj and Robert Stahelin and Langlois, {Robert E.} and Wonhwa Cho and Hui Lu",
year = "2006",
month = "6",
day = "2",
doi = "10.1016/j.jmb.2006.03.039",
language = "English (US)",
volume = "359",
pages = "486--495",
journal = "Journal of Molecular Biology",
issn = "0022-2836",
publisher = "Academic Press Inc.",
number = "2",

}

TY - JOUR

T1 - Structural Bioinformatics Prediction of Membrane-binding Proteins

AU - Bhardwaj, Nitin

AU - Stahelin, Robert

AU - Langlois, Robert E.

AU - Cho, Wonhwa

AU - Lu, Hui

PY - 2006/6/2

Y1 - 2006/6/2

N2 - Membrane-binding peripheral proteins play important roles in many biological processes, including cell signaling and membrane trafficking. Unlike integral membrane proteins, these proteins bind the membrane mostly in a reversible manner. Since peripheral proteins do not have canonical transmembrane segments, it is difficult to identify them from their amino acid sequences. As a first step toward genome-scale identification of membrane-binding peripheral proteins, we built a kernel-based machine learning protocol. Key features of known membrane-binding proteins, including electrostatic properties and amino acid composition, were calculated from their amino acid sequences and tertiary structures, which were then incorporated into the support vector machine to perform the classification. A data set of 40 membrane-binding proteins and 230 non-membrane-binding proteins was used to construct and validate the protocol. Cross-validation and holdout evaluation of the protocol showed that the accuracy of the prediction reached up to 93.7% and 91.6%, respectively. The protocol was applied to the prediction of membrane-binding properties of four C2 domains from novel protein kinases C. Although these C2 domains have 50% sequence identity, only one of them was predicted to bind the membrane, which was verified experimentally with surface plasmon resonance analysis. These results suggest that our protocol can be used for predicting membrane-binding properties of a wide variety of modular domains and may be further extended to genome-scale identification of membrane-binding peripheral proteins.

AB - Membrane-binding peripheral proteins play important roles in many biological processes, including cell signaling and membrane trafficking. Unlike integral membrane proteins, these proteins bind the membrane mostly in a reversible manner. Since peripheral proteins do not have canonical transmembrane segments, it is difficult to identify them from their amino acid sequences. As a first step toward genome-scale identification of membrane-binding peripheral proteins, we built a kernel-based machine learning protocol. Key features of known membrane-binding proteins, including electrostatic properties and amino acid composition, were calculated from their amino acid sequences and tertiary structures, which were then incorporated into the support vector machine to perform the classification. A data set of 40 membrane-binding proteins and 230 non-membrane-binding proteins was used to construct and validate the protocol. Cross-validation and holdout evaluation of the protocol showed that the accuracy of the prediction reached up to 93.7% and 91.6%, respectively. The protocol was applied to the prediction of membrane-binding properties of four C2 domains from novel protein kinases C. Although these C2 domains have 50% sequence identity, only one of them was predicted to bind the membrane, which was verified experimentally with surface plasmon resonance analysis. These results suggest that our protocol can be used for predicting membrane-binding properties of a wide variety of modular domains and may be further extended to genome-scale identification of membrane-binding peripheral proteins.

KW - function annotation

KW - peripheral proteins

KW - protein function prediction

KW - protein-membrane interactions

KW - support vector machines

UR - http://www.scopus.com/inward/record.url?scp=33646192301&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33646192301&partnerID=8YFLogxK

U2 - 10.1016/j.jmb.2006.03.039

DO - 10.1016/j.jmb.2006.03.039

M3 - Article

C2 - 16626739

AN - SCOPUS:33646192301

VL - 359

SP - 486

EP - 495

JO - Journal of Molecular Biology

JF - Journal of Molecular Biology

SN - 0022-2836

IS - 2

ER -