Empirical bayes conditional independence graphs for regulatory network recovery

Rami Mahdi, Abishek S. Madduri, Guoqing Wang, Yael Strulovici-barel, Jacqueline Salit, Neil R. Hackett, Ronald Crystal, Jason G. Mezey

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Motivation: Computational inference methods that make use of graphical models to extract regulatory networks from gene expression data can have difficulty reconstructing dense regions of a network, a consequence of both computational complexity and unreliable parameter estimation when sample size is small. As a result, identification of hub genes is of special difficulty for these methods. Methods: We present a new algorithm, Empirical Light Mutual Min (ELMM), for large network reconstruction that has properties well suited for recovery of graphs with high-degree nodes. ELMM reconstructs the undirected graph of a regulatory network using empirical Bayes conditional independence testing with a heuristic relaxation of independence constraints in dense areas of the graph. This relaxation allows only one gene of a pair with a putative relation to be aware of the network connection, an approach that is aimed at easing multiple testing problems associated with recovering densely connected structures. Results: Using in silico data, we show that ELMM has better performance than commonly used network inference algorithms including GeneNet, ARACNE, FOCI, GENIE3 and GLASSO. We also apply ELMM to reconstruct a network among 5492 genes expressed in human lung airway epithelium of healthy non-smokers, healthy smokers and individuals with chronic obstructive pulmonary disease assayed using microarrays. The analysis identifies dense sub-networks that are consistent with known regulatory relationships in the lung airway and also suggests novel hub regulatory relationships among a number of genes that play roles in oxidative stress and secretion.

Original languageEnglish
Article numberbts312
Pages (from-to)2029-2036
Number of pages8
JournalBioinformatics
Volume28
Issue number15
DOIs
Publication statusPublished - 1 Aug 2012
Externally publishedYes

Fingerprint

Empirical Bayes
Conditional Independence
Regulatory Networks
Recovery
Genes
Light
Graph in graph theory
Gene
Lung
Pulmonary diseases
Oxidative stress
Testing
Microarrays
Gene expression
Computer Simulation
Sample Size
Parameter estimation
Oxidative Stress
Chronic Obstructive Pulmonary Disease
Computational complexity

ASJC Scopus subject areas

  • Biochemistry
  • Molecular Biology
  • Computational Theory and Mathematics
  • Computer Science Applications
  • Computational Mathematics
  • Statistics and Probability
  • Medicine(all)

Cite this

Mahdi, R., Madduri, A. S., Wang, G., Strulovici-barel, Y., Salit, J., Hackett, N. R., ... Mezey, J. G. (2012). Empirical bayes conditional independence graphs for regulatory network recovery. Bioinformatics, 28(15), 2029-2036. [bts312]. https://doi.org/10.1093/bioinformatics/bts312

Empirical bayes conditional independence graphs for regulatory network recovery. / Mahdi, Rami; Madduri, Abishek S.; Wang, Guoqing; Strulovici-barel, Yael; Salit, Jacqueline; Hackett, Neil R.; Crystal, Ronald; Mezey, Jason G.

In: Bioinformatics, Vol. 28, No. 15, bts312, 01.08.2012, p. 2029-2036.

Research output: Contribution to journalArticle

Mahdi, R, Madduri, AS, Wang, G, Strulovici-barel, Y, Salit, J, Hackett, NR, Crystal, R & Mezey, JG 2012, 'Empirical bayes conditional independence graphs for regulatory network recovery', Bioinformatics, vol. 28, no. 15, bts312, pp. 2029-2036. https://doi.org/10.1093/bioinformatics/bts312
Mahdi R, Madduri AS, Wang G, Strulovici-barel Y, Salit J, Hackett NR et al. Empirical bayes conditional independence graphs for regulatory network recovery. Bioinformatics. 2012 Aug 1;28(15):2029-2036. bts312. https://doi.org/10.1093/bioinformatics/bts312
Mahdi, Rami ; Madduri, Abishek S. ; Wang, Guoqing ; Strulovici-barel, Yael ; Salit, Jacqueline ; Hackett, Neil R. ; Crystal, Ronald ; Mezey, Jason G. / Empirical bayes conditional independence graphs for regulatory network recovery. In: Bioinformatics. 2012 ; Vol. 28, No. 15. pp. 2029-2036.
@article{b66909f45e484f5b87b60018dc202df1,
title = "Empirical bayes conditional independence graphs for regulatory network recovery",
abstract = "Motivation: Computational inference methods that make use of graphical models to extract regulatory networks from gene expression data can have difficulty reconstructing dense regions of a network, a consequence of both computational complexity and unreliable parameter estimation when sample size is small. As a result, identification of hub genes is of special difficulty for these methods. Methods: We present a new algorithm, Empirical Light Mutual Min (ELMM), for large network reconstruction that has properties well suited for recovery of graphs with high-degree nodes. ELMM reconstructs the undirected graph of a regulatory network using empirical Bayes conditional independence testing with a heuristic relaxation of independence constraints in dense areas of the graph. This relaxation allows only one gene of a pair with a putative relation to be aware of the network connection, an approach that is aimed at easing multiple testing problems associated with recovering densely connected structures. Results: Using in silico data, we show that ELMM has better performance than commonly used network inference algorithms including GeneNet, ARACNE, FOCI, GENIE3 and GLASSO. We also apply ELMM to reconstruct a network among 5492 genes expressed in human lung airway epithelium of healthy non-smokers, healthy smokers and individuals with chronic obstructive pulmonary disease assayed using microarrays. The analysis identifies dense sub-networks that are consistent with known regulatory relationships in the lung airway and also suggests novel hub regulatory relationships among a number of genes that play roles in oxidative stress and secretion.",
author = "Rami Mahdi and Madduri, {Abishek S.} and Guoqing Wang and Yael Strulovici-barel and Jacqueline Salit and Hackett, {Neil R.} and Ronald Crystal and Mezey, {Jason G.}",
year = "2012",
month = "8",
day = "1",
doi = "10.1093/bioinformatics/bts312",
language = "English",
volume = "28",
pages = "2029--2036",
journal = "Bioinformatics",
issn = "1367-4803",
publisher = "Oxford University Press",
number = "15",

}

TY - JOUR

T1 - Empirical bayes conditional independence graphs for regulatory network recovery

AU - Mahdi, Rami

AU - Madduri, Abishek S.

AU - Wang, Guoqing

AU - Strulovici-barel, Yael

AU - Salit, Jacqueline

AU - Hackett, Neil R.

AU - Crystal, Ronald

AU - Mezey, Jason G.

PY - 2012/8/1

Y1 - 2012/8/1

N2 - Motivation: Computational inference methods that make use of graphical models to extract regulatory networks from gene expression data can have difficulty reconstructing dense regions of a network, a consequence of both computational complexity and unreliable parameter estimation when sample size is small. As a result, identification of hub genes is of special difficulty for these methods. Methods: We present a new algorithm, Empirical Light Mutual Min (ELMM), for large network reconstruction that has properties well suited for recovery of graphs with high-degree nodes. ELMM reconstructs the undirected graph of a regulatory network using empirical Bayes conditional independence testing with a heuristic relaxation of independence constraints in dense areas of the graph. This relaxation allows only one gene of a pair with a putative relation to be aware of the network connection, an approach that is aimed at easing multiple testing problems associated with recovering densely connected structures. Results: Using in silico data, we show that ELMM has better performance than commonly used network inference algorithms including GeneNet, ARACNE, FOCI, GENIE3 and GLASSO. We also apply ELMM to reconstruct a network among 5492 genes expressed in human lung airway epithelium of healthy non-smokers, healthy smokers and individuals with chronic obstructive pulmonary disease assayed using microarrays. The analysis identifies dense sub-networks that are consistent with known regulatory relationships in the lung airway and also suggests novel hub regulatory relationships among a number of genes that play roles in oxidative stress and secretion.

AB - Motivation: Computational inference methods that make use of graphical models to extract regulatory networks from gene expression data can have difficulty reconstructing dense regions of a network, a consequence of both computational complexity and unreliable parameter estimation when sample size is small. As a result, identification of hub genes is of special difficulty for these methods. Methods: We present a new algorithm, Empirical Light Mutual Min (ELMM), for large network reconstruction that has properties well suited for recovery of graphs with high-degree nodes. ELMM reconstructs the undirected graph of a regulatory network using empirical Bayes conditional independence testing with a heuristic relaxation of independence constraints in dense areas of the graph. This relaxation allows only one gene of a pair with a putative relation to be aware of the network connection, an approach that is aimed at easing multiple testing problems associated with recovering densely connected structures. Results: Using in silico data, we show that ELMM has better performance than commonly used network inference algorithms including GeneNet, ARACNE, FOCI, GENIE3 and GLASSO. We also apply ELMM to reconstruct a network among 5492 genes expressed in human lung airway epithelium of healthy non-smokers, healthy smokers and individuals with chronic obstructive pulmonary disease assayed using microarrays. The analysis identifies dense sub-networks that are consistent with known regulatory relationships in the lung airway and also suggests novel hub regulatory relationships among a number of genes that play roles in oxidative stress and secretion.

UR - http://www.scopus.com/inward/record.url?scp=84865153246&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84865153246&partnerID=8YFLogxK

U2 - 10.1093/bioinformatics/bts312

DO - 10.1093/bioinformatics/bts312

M3 - Article

C2 - 22685074

AN - SCOPUS:84865153246

VL - 28

SP - 2029

EP - 2036

JO - Bioinformatics

JF - Bioinformatics

SN - 1367-4803

IS - 15

M1 - bts312

ER -