Sparse Inverse Covariance Estimation with L0 Penalty for Network Construction with Omics Data

Zhenqiu Liu, Shili Lin, Nan Deng, Dermot P B McGovern, Steven Piantadosi

Research output: Contribution to journalArticle

Abstract

Constructing coexpression and association networks with omics data is crucial for studying gene-gene interactions and underlying biological mechanisms. In recent years, learning the structure of a Gaussian graphical model from high-dimensional data using L1 penalty has been well-studied and many applications in bioinformatics and computational biology have been found. However, besides the problem of biased estimators with LASSO, L1 does not always choose the true model consistently. Based on our previous work with L0 regularized regression (Liu and Li, 2014), we propose an L0 regularized sparse inverse covariance estimation (L0RICE) for structure learning with the efficient alternating direction (AD) method. The proposed method is robust and has the oracle property. The proposed method is applied to omics data including data, from next-generation sequencing technologies. Novel procedures for network construction and high-order gene-gene interaction detection with omics data are developed. Results from simulation and real omics data analysis indicate that L0 regularized structure learning can identify high-order correlation structure with lower false positive rate and outperform graphical lasso by a large margin.

Original languageEnglish (US)
Pages (from-to)192-202
Number of pages11
JournalJournal of Computational Biology
Volume23
Issue number3
DOIs
StatePublished - Mar 1 2016
Externally publishedYes

Fingerprint

Covariance Estimation
Penalty
Genes
Gene
Structure Learning
Learning
Computational Biology
Higher Order
Oracle Property
Alternating Direction Method
Lasso
Gene Order
Correlation Structure
Gaussian Model
Graphical Models
High-dimensional Data
Bioinformatics
False Positive
Interaction
Margin

Keywords

  • algorithms
  • graphs and networks
  • haplotypes
  • machine learning
  • metagenomics
  • statistical models

ASJC Scopus subject areas

  • Molecular Biology
  • Genetics
  • Computational Mathematics
  • Modeling and Simulation
  • Computational Theory and Mathematics

Cite this

Sparse Inverse Covariance Estimation with L0 Penalty for Network Construction with Omics Data. / Liu, Zhenqiu; Lin, Shili; Deng, Nan; McGovern, Dermot P B; Piantadosi, Steven.

In: Journal of Computational Biology, Vol. 23, No. 3, 01.03.2016, p. 192-202.

Research output: Contribution to journalArticle

Liu, Zhenqiu ; Lin, Shili ; Deng, Nan ; McGovern, Dermot P B ; Piantadosi, Steven. / Sparse Inverse Covariance Estimation with L0 Penalty for Network Construction with Omics Data. In: Journal of Computational Biology. 2016 ; Vol. 23, No. 3. pp. 192-202.
@article{3b0d3018c60e4954bab6ad4bb87f839a,
title = "Sparse Inverse Covariance Estimation with L0 Penalty for Network Construction with Omics Data",
abstract = "Constructing coexpression and association networks with omics data is crucial for studying gene-gene interactions and underlying biological mechanisms. In recent years, learning the structure of a Gaussian graphical model from high-dimensional data using L1 penalty has been well-studied and many applications in bioinformatics and computational biology have been found. However, besides the problem of biased estimators with LASSO, L1 does not always choose the true model consistently. Based on our previous work with L0 regularized regression (Liu and Li, 2014), we propose an L0 regularized sparse inverse covariance estimation (L0RICE) for structure learning with the efficient alternating direction (AD) method. The proposed method is robust and has the oracle property. The proposed method is applied to omics data including data, from next-generation sequencing technologies. Novel procedures for network construction and high-order gene-gene interaction detection with omics data are developed. Results from simulation and real omics data analysis indicate that L0 regularized structure learning can identify high-order correlation structure with lower false positive rate and outperform graphical lasso by a large margin.",
keywords = "algorithms, graphs and networks, haplotypes, machine learning, metagenomics, statistical models",
author = "Zhenqiu Liu and Shili Lin and Nan Deng and McGovern, {Dermot P B} and Steven Piantadosi",
year = "2016",
month = "3",
day = "1",
doi = "10.1089/cmb.2015.0102",
language = "English (US)",
volume = "23",
pages = "192--202",
journal = "Journal of Computational Biology",
issn = "1066-5277",
publisher = "Mary Ann Liebert Inc.",
number = "3",

}

TY - JOUR

T1 - Sparse Inverse Covariance Estimation with L0 Penalty for Network Construction with Omics Data

AU - Liu, Zhenqiu

AU - Lin, Shili

AU - Deng, Nan

AU - McGovern, Dermot P B

AU - Piantadosi, Steven

PY - 2016/3/1

Y1 - 2016/3/1

N2 - Constructing coexpression and association networks with omics data is crucial for studying gene-gene interactions and underlying biological mechanisms. In recent years, learning the structure of a Gaussian graphical model from high-dimensional data using L1 penalty has been well-studied and many applications in bioinformatics and computational biology have been found. However, besides the problem of biased estimators with LASSO, L1 does not always choose the true model consistently. Based on our previous work with L0 regularized regression (Liu and Li, 2014), we propose an L0 regularized sparse inverse covariance estimation (L0RICE) for structure learning with the efficient alternating direction (AD) method. The proposed method is robust and has the oracle property. The proposed method is applied to omics data including data, from next-generation sequencing technologies. Novel procedures for network construction and high-order gene-gene interaction detection with omics data are developed. Results from simulation and real omics data analysis indicate that L0 regularized structure learning can identify high-order correlation structure with lower false positive rate and outperform graphical lasso by a large margin.

AB - Constructing coexpression and association networks with omics data is crucial for studying gene-gene interactions and underlying biological mechanisms. In recent years, learning the structure of a Gaussian graphical model from high-dimensional data using L1 penalty has been well-studied and many applications in bioinformatics and computational biology have been found. However, besides the problem of biased estimators with LASSO, L1 does not always choose the true model consistently. Based on our previous work with L0 regularized regression (Liu and Li, 2014), we propose an L0 regularized sparse inverse covariance estimation (L0RICE) for structure learning with the efficient alternating direction (AD) method. The proposed method is robust and has the oracle property. The proposed method is applied to omics data including data, from next-generation sequencing technologies. Novel procedures for network construction and high-order gene-gene interaction detection with omics data are developed. Results from simulation and real omics data analysis indicate that L0 regularized structure learning can identify high-order correlation structure with lower false positive rate and outperform graphical lasso by a large margin.

KW - algorithms

KW - graphs and networks

KW - haplotypes

KW - machine learning

KW - metagenomics

KW - statistical models

UR - http://www.scopus.com/inward/record.url?scp=84960884492&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84960884492&partnerID=8YFLogxK

U2 - 10.1089/cmb.2015.0102

DO - 10.1089/cmb.2015.0102

M3 - Article

VL - 23

SP - 192

EP - 202

JO - Journal of Computational Biology

JF - Journal of Computational Biology

SN - 1066-5277

IS - 3

ER -