Reconstruct modular phenotype-specific gene networks by knowledge-driven matrix factorization

Xuerui Yang, Yang Zhou, Rong Jin, Christina Chan

Research output: Contribution to journalArticle

  • 15 Citations

Abstract

Motivation: Reconstructing gene networks from microarray data has provided mechanistic information on cellular processes. A popular structure learning method, Bayesian network inference, has been used to determine network topology despite its shortcomings, i.e. the high-computational cost when analyzing a large number of genes and the inefficiency in exploiting prior knowledge, such as the co-regulation information of the genes. To address these limitations, we are introducing an alternative method, knowledge-driven matrix factorization (KMF) framework, to reconstruct phenotype-specific modular gene networks. Results: Considering the reconstruction of gene network as a matrix factorization problem, we first use the gene expression data to estimate a correlation matrix, and then factorize the correlation matrix to recover the gene modules and the interactions between them. Prior knowledge from Gene Ontology is integrated into the matrix factorization. We applied this KMF algorithm to hepatocellular carcinoma (HepG2) cells treated with free fatty acids (FFAs). By comparing the module networks for the different conditions, we identified the specific modules that are involved in conferring the cytotoxic phenotype induced by palmitate. Further analysis of the gene modules of the different conditions suggested individual genes that play important roles in palmitate-induced cytotoxicity. In summary, KMF can efficiently integrate gene expression data with prior knowledge, thereby providing a powerful method of reconstructing phenotype-specific gene networks and valuable insights into the mechanisms that govern the phenotype.

LanguageEnglish (US)
Pages2236-2243
Number of pages8
JournalBioinformatics
Volume25
Issue number17
DOIs
StatePublished - 2009

Profile

Gene Networks
Matrix Factorization
Gene Regulatory Networks
Factorization
Phenotype
Genes
Gene
Prior Knowledge
Module
Correlation Matrix
Gene Expression Data
Palmitates
Factorise
Structure Learning
Cytotoxicity
Gene Ontology
Fatty Acids
Gene Expression
Microarray Data
Bayesian Networks

ASJC Scopus subject areas

  • Biochemistry
  • Molecular Biology
  • Computational Theory and Mathematics
  • Computer Science Applications
  • Computational Mathematics
  • Statistics and Probability

Cite this

Reconstruct modular phenotype-specific gene networks by knowledge-driven matrix factorization. / Yang, Xuerui; Zhou, Yang; Jin, Rong; Chan, Christina.

In: Bioinformatics, Vol. 25, No. 17, 2009, p. 2236-2243.

Research output: Contribution to journalArticle

Yang, Xuerui ; Zhou, Yang ; Jin, Rong ; Chan, Christina. / Reconstruct modular phenotype-specific gene networks by knowledge-driven matrix factorization. In: Bioinformatics. 2009 ; Vol. 25, No. 17. pp. 2236-2243
@article{c98fe1753e4b448890d5541504fc635a,
title = "Reconstruct modular phenotype-specific gene networks by knowledge-driven matrix factorization",
abstract = "Motivation: Reconstructing gene networks from microarray data has provided mechanistic information on cellular processes. A popular structure learning method, Bayesian network inference, has been used to determine network topology despite its shortcomings, i.e. the high-computational cost when analyzing a large number of genes and the inefficiency in exploiting prior knowledge, such as the co-regulation information of the genes. To address these limitations, we are introducing an alternative method, knowledge-driven matrix factorization (KMF) framework, to reconstruct phenotype-specific modular gene networks. Results: Considering the reconstruction of gene network as a matrix factorization problem, we first use the gene expression data to estimate a correlation matrix, and then factorize the correlation matrix to recover the gene modules and the interactions between them. Prior knowledge from Gene Ontology is integrated into the matrix factorization. We applied this KMF algorithm to hepatocellular carcinoma (HepG2) cells treated with free fatty acids (FFAs). By comparing the module networks for the different conditions, we identified the specific modules that are involved in conferring the cytotoxic phenotype induced by palmitate. Further analysis of the gene modules of the different conditions suggested individual genes that play important roles in palmitate-induced cytotoxicity. In summary, KMF can efficiently integrate gene expression data with prior knowledge, thereby providing a powerful method of reconstructing phenotype-specific gene networks and valuable insights into the mechanisms that govern the phenotype.",
author = "Xuerui Yang and Yang Zhou and Rong Jin and Christina Chan",
year = "2009",
doi = "10.1093/bioinformatics/btp376",
language = "English (US)",
volume = "25",
pages = "2236--2243",
journal = "Bioinformatics",
issn = "1367-4803",
publisher = "Oxford University Press",
number = "17",

}

TY - JOUR

T1 - Reconstruct modular phenotype-specific gene networks by knowledge-driven matrix factorization

AU - Yang,Xuerui

AU - Zhou,Yang

AU - Jin,Rong

AU - Chan,Christina

PY - 2009

Y1 - 2009

N2 - Motivation: Reconstructing gene networks from microarray data has provided mechanistic information on cellular processes. A popular structure learning method, Bayesian network inference, has been used to determine network topology despite its shortcomings, i.e. the high-computational cost when analyzing a large number of genes and the inefficiency in exploiting prior knowledge, such as the co-regulation information of the genes. To address these limitations, we are introducing an alternative method, knowledge-driven matrix factorization (KMF) framework, to reconstruct phenotype-specific modular gene networks. Results: Considering the reconstruction of gene network as a matrix factorization problem, we first use the gene expression data to estimate a correlation matrix, and then factorize the correlation matrix to recover the gene modules and the interactions between them. Prior knowledge from Gene Ontology is integrated into the matrix factorization. We applied this KMF algorithm to hepatocellular carcinoma (HepG2) cells treated with free fatty acids (FFAs). By comparing the module networks for the different conditions, we identified the specific modules that are involved in conferring the cytotoxic phenotype induced by palmitate. Further analysis of the gene modules of the different conditions suggested individual genes that play important roles in palmitate-induced cytotoxicity. In summary, KMF can efficiently integrate gene expression data with prior knowledge, thereby providing a powerful method of reconstructing phenotype-specific gene networks and valuable insights into the mechanisms that govern the phenotype.

AB - Motivation: Reconstructing gene networks from microarray data has provided mechanistic information on cellular processes. A popular structure learning method, Bayesian network inference, has been used to determine network topology despite its shortcomings, i.e. the high-computational cost when analyzing a large number of genes and the inefficiency in exploiting prior knowledge, such as the co-regulation information of the genes. To address these limitations, we are introducing an alternative method, knowledge-driven matrix factorization (KMF) framework, to reconstruct phenotype-specific modular gene networks. Results: Considering the reconstruction of gene network as a matrix factorization problem, we first use the gene expression data to estimate a correlation matrix, and then factorize the correlation matrix to recover the gene modules and the interactions between them. Prior knowledge from Gene Ontology is integrated into the matrix factorization. We applied this KMF algorithm to hepatocellular carcinoma (HepG2) cells treated with free fatty acids (FFAs). By comparing the module networks for the different conditions, we identified the specific modules that are involved in conferring the cytotoxic phenotype induced by palmitate. Further analysis of the gene modules of the different conditions suggested individual genes that play important roles in palmitate-induced cytotoxicity. In summary, KMF can efficiently integrate gene expression data with prior knowledge, thereby providing a powerful method of reconstructing phenotype-specific gene networks and valuable insights into the mechanisms that govern the phenotype.

UR - http://www.scopus.com/inward/record.url?scp=69949134573&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=69949134573&partnerID=8YFLogxK

U2 - 10.1093/bioinformatics/btp376

DO - 10.1093/bioinformatics/btp376

M3 - Article

VL - 25

SP - 2236

EP - 2243

JO - Bioinformatics

T2 - Bioinformatics

JF - Bioinformatics

SN - 1367-4803

IS - 17

ER -