Gene function prediction with gene interaction networks: A context graph kernel approach

Xin Li, Hsinchun Chen, Jiexun Li, Zhu Zhang

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

Predicting gene functions is a challenge for biologists in the postgenomic era. Interactions among genes and their products compose networks that can be used to infer gene functions. Most previous studies adopt a linkage assumption, i.e., they assume that gene interactions indicate functional similarities between connected genes. In this study, we propose to use a gene's context graph, i.e., the gene interaction network associated with the focal gene, to infer its functions. In a kernel-based machine-learning framework, we design a context graph kernel to capture the information in context graphs. Our experimental study on a testbed of p53-related genes demonstrates the advantage of using indirect gene interactions and shows the empirical superiority of the proposed approach over linkage-assumption-based methods, such as the algorithm to minimize inconsistent connected genes and diffusion kernels.

Original languageEnglish (US)
Article number5272443
Pages (from-to)119-128
Number of pages10
JournalIEEE Transactions on Information Technology in Biomedicine
Volume14
Issue number1
DOIs
StatePublished - Jan 2010

Fingerprint

Gene Regulatory Networks
Genes
p53 Genes
Testbeds
Learning systems

Keywords

  • Classification
  • Gene pathway
  • Kernel-based method
  • System biology

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Biotechnology
  • Computer Science Applications
  • Medicine(all)

Cite this

Gene function prediction with gene interaction networks : A context graph kernel approach. / Li, Xin; Chen, Hsinchun; Li, Jiexun; Zhang, Zhu.

In: IEEE Transactions on Information Technology in Biomedicine, Vol. 14, No. 1, 5272443, 01.2010, p. 119-128.

Research output: Contribution to journalArticle

@article{7c396f8e40bf4874b6a2270523d2d261,
title = "Gene function prediction with gene interaction networks: A context graph kernel approach",
abstract = "Predicting gene functions is a challenge for biologists in the postgenomic era. Interactions among genes and their products compose networks that can be used to infer gene functions. Most previous studies adopt a linkage assumption, i.e., they assume that gene interactions indicate functional similarities between connected genes. In this study, we propose to use a gene's context graph, i.e., the gene interaction network associated with the focal gene, to infer its functions. In a kernel-based machine-learning framework, we design a context graph kernel to capture the information in context graphs. Our experimental study on a testbed of p53-related genes demonstrates the advantage of using indirect gene interactions and shows the empirical superiority of the proposed approach over linkage-assumption-based methods, such as the algorithm to minimize inconsistent connected genes and diffusion kernels.",
keywords = "Classification, Gene pathway, Kernel-based method, System biology",
author = "Xin Li and Hsinchun Chen and Jiexun Li and Zhu Zhang",
year = "2010",
month = "1",
doi = "10.1109/TITB.2009.2033116",
language = "English (US)",
volume = "14",
pages = "119--128",
journal = "IEEE Journal of Biomedical and Health Informatics",
issn = "2168-2194",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "1",

}

TY - JOUR

T1 - Gene function prediction with gene interaction networks

T2 - A context graph kernel approach

AU - Li, Xin

AU - Chen, Hsinchun

AU - Li, Jiexun

AU - Zhang, Zhu

PY - 2010/1

Y1 - 2010/1

N2 - Predicting gene functions is a challenge for biologists in the postgenomic era. Interactions among genes and their products compose networks that can be used to infer gene functions. Most previous studies adopt a linkage assumption, i.e., they assume that gene interactions indicate functional similarities between connected genes. In this study, we propose to use a gene's context graph, i.e., the gene interaction network associated with the focal gene, to infer its functions. In a kernel-based machine-learning framework, we design a context graph kernel to capture the information in context graphs. Our experimental study on a testbed of p53-related genes demonstrates the advantage of using indirect gene interactions and shows the empirical superiority of the proposed approach over linkage-assumption-based methods, such as the algorithm to minimize inconsistent connected genes and diffusion kernels.

AB - Predicting gene functions is a challenge for biologists in the postgenomic era. Interactions among genes and their products compose networks that can be used to infer gene functions. Most previous studies adopt a linkage assumption, i.e., they assume that gene interactions indicate functional similarities between connected genes. In this study, we propose to use a gene's context graph, i.e., the gene interaction network associated with the focal gene, to infer its functions. In a kernel-based machine-learning framework, we design a context graph kernel to capture the information in context graphs. Our experimental study on a testbed of p53-related genes demonstrates the advantage of using indirect gene interactions and shows the empirical superiority of the proposed approach over linkage-assumption-based methods, such as the algorithm to minimize inconsistent connected genes and diffusion kernels.

KW - Classification

KW - Gene pathway

KW - Kernel-based method

KW - System biology

UR - http://www.scopus.com/inward/record.url?scp=76849095353&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=76849095353&partnerID=8YFLogxK

U2 - 10.1109/TITB.2009.2033116

DO - 10.1109/TITB.2009.2033116

M3 - Article

C2 - 19789115

AN - SCOPUS:76849095353

VL - 14

SP - 119

EP - 128

JO - IEEE Journal of Biomedical and Health Informatics

JF - IEEE Journal of Biomedical and Health Informatics

SN - 2168-2194

IS - 1

M1 - 5272443

ER -