Computational analysis of muscular dystrophy sub-types using a novel integrative scheme

Chen Wang, Sook Ha, Jianhua Xuan, Yue Wang, Eric Hoffman

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

To construct biologically interpretable gene sets for muscular dystrophy (MD) sub-type classification, we propose a novel computational scheme to integrate protein-protein interaction (PPI) network, functional gene set information, and mRNA profiling data. The workflow of the proposed scheme includes the following three major steps: firstly, we apply an affinity propagation clustering (APC) approach to identify gene sub-networks associated with each MD sub-type, in which a new distance metric is proposed for APC to combine PPI network information and gene-gene co-expression relationship; secondly, we further incorporate functional gene set knowledge, which complements the physical PPI information, into our scheme for biomarker identification; finally, based on the constructed sub-networks and gene set features, we apply multiclass support vector machines (MSVMs) for MD sub-type classification, with which to highlight the biomarkers contributing to sub-type prediction. The experimental results show that our scheme can help identify sub-networks and gene sets that are more relevant to MD than those constructed by other conventional approaches. Moreover, our integrative strategy improves the prediction accuracy substantially, especially for those 'hard-to-classify' sub-types.

Original languageEnglish (US)
Pages (from-to)9-17
Number of pages9
JournalNeurocomputing
Volume92
DOIs
StatePublished - Sep 1 2012
Externally publishedYes

Fingerprint

Muscular Dystrophies
Gene Regulatory Networks
Genes
Protein Interaction Maps
Proteins
Cluster Analysis
Biomarkers
Workflow
Complement System Proteins
Gene Expression
Messenger RNA
Support vector machines

Keywords

  • Affinity propagation clustering
  • Biomarker discovery
  • Classification
  • Gene expression
  • Muscular dystrophy

ASJC Scopus subject areas

  • Computer Science Applications
  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this

Computational analysis of muscular dystrophy sub-types using a novel integrative scheme. / Wang, Chen; Ha, Sook; Xuan, Jianhua; Wang, Yue; Hoffman, Eric.

In: Neurocomputing, Vol. 92, 01.09.2012, p. 9-17.

Research output: Contribution to journalArticle

Wang, Chen ; Ha, Sook ; Xuan, Jianhua ; Wang, Yue ; Hoffman, Eric. / Computational analysis of muscular dystrophy sub-types using a novel integrative scheme. In: Neurocomputing. 2012 ; Vol. 92. pp. 9-17.
@article{d88229e8c296444381dbb23a40a33431,
title = "Computational analysis of muscular dystrophy sub-types using a novel integrative scheme",
abstract = "To construct biologically interpretable gene sets for muscular dystrophy (MD) sub-type classification, we propose a novel computational scheme to integrate protein-protein interaction (PPI) network, functional gene set information, and mRNA profiling data. The workflow of the proposed scheme includes the following three major steps: firstly, we apply an affinity propagation clustering (APC) approach to identify gene sub-networks associated with each MD sub-type, in which a new distance metric is proposed for APC to combine PPI network information and gene-gene co-expression relationship; secondly, we further incorporate functional gene set knowledge, which complements the physical PPI information, into our scheme for biomarker identification; finally, based on the constructed sub-networks and gene set features, we apply multiclass support vector machines (MSVMs) for MD sub-type classification, with which to highlight the biomarkers contributing to sub-type prediction. The experimental results show that our scheme can help identify sub-networks and gene sets that are more relevant to MD than those constructed by other conventional approaches. Moreover, our integrative strategy improves the prediction accuracy substantially, especially for those 'hard-to-classify' sub-types.",
keywords = "Affinity propagation clustering, Biomarker discovery, Classification, Gene expression, Muscular dystrophy",
author = "Chen Wang and Sook Ha and Jianhua Xuan and Yue Wang and Eric Hoffman",
year = "2012",
month = "9",
day = "1",
doi = "10.1016/j.neucom.2011.08.037",
language = "English (US)",
volume = "92",
pages = "9--17",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",

}

TY - JOUR

T1 - Computational analysis of muscular dystrophy sub-types using a novel integrative scheme

AU - Wang, Chen

AU - Ha, Sook

AU - Xuan, Jianhua

AU - Wang, Yue

AU - Hoffman, Eric

PY - 2012/9/1

Y1 - 2012/9/1

N2 - To construct biologically interpretable gene sets for muscular dystrophy (MD) sub-type classification, we propose a novel computational scheme to integrate protein-protein interaction (PPI) network, functional gene set information, and mRNA profiling data. The workflow of the proposed scheme includes the following three major steps: firstly, we apply an affinity propagation clustering (APC) approach to identify gene sub-networks associated with each MD sub-type, in which a new distance metric is proposed for APC to combine PPI network information and gene-gene co-expression relationship; secondly, we further incorporate functional gene set knowledge, which complements the physical PPI information, into our scheme for biomarker identification; finally, based on the constructed sub-networks and gene set features, we apply multiclass support vector machines (MSVMs) for MD sub-type classification, with which to highlight the biomarkers contributing to sub-type prediction. The experimental results show that our scheme can help identify sub-networks and gene sets that are more relevant to MD than those constructed by other conventional approaches. Moreover, our integrative strategy improves the prediction accuracy substantially, especially for those 'hard-to-classify' sub-types.

AB - To construct biologically interpretable gene sets for muscular dystrophy (MD) sub-type classification, we propose a novel computational scheme to integrate protein-protein interaction (PPI) network, functional gene set information, and mRNA profiling data. The workflow of the proposed scheme includes the following three major steps: firstly, we apply an affinity propagation clustering (APC) approach to identify gene sub-networks associated with each MD sub-type, in which a new distance metric is proposed for APC to combine PPI network information and gene-gene co-expression relationship; secondly, we further incorporate functional gene set knowledge, which complements the physical PPI information, into our scheme for biomarker identification; finally, based on the constructed sub-networks and gene set features, we apply multiclass support vector machines (MSVMs) for MD sub-type classification, with which to highlight the biomarkers contributing to sub-type prediction. The experimental results show that our scheme can help identify sub-networks and gene sets that are more relevant to MD than those constructed by other conventional approaches. Moreover, our integrative strategy improves the prediction accuracy substantially, especially for those 'hard-to-classify' sub-types.

KW - Affinity propagation clustering

KW - Biomarker discovery

KW - Classification

KW - Gene expression

KW - Muscular dystrophy

UR - http://www.scopus.com/inward/record.url?scp=84862784886&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862784886&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2011.08.037

DO - 10.1016/j.neucom.2011.08.037

M3 - Article

AN - SCOPUS:84862784886

VL - 92

SP - 9

EP - 17

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

ER -