Ridge estimation of network models from time-course omics data

Viktorian Miok, Saskia M. Wilting, Wessel N. van Wieringen

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

Time-course omics experiments enable the reconstruction of the dynamics of the cellular regulatory network. Here, we describe the means for this reconstruction and the downstream exploitation of the inferred network. It is assumed that one of the various vector-autoregressive models (VAR) models presented here serves as a reasonably accurate description of the time-course omics data. The models are estimated through ridge penalized likelihood maximization, accompanied by functionality for the determination of optimal penalty paramaters. Prior knowledge on the network topology is accommodated by the estimation procedures. Various routes that translate the fitted models into more tangible implications for the medical researcher are described. The network is inferred from the—nonsparse—ridge estimates through empirical Bayes probabilistic thresholding. The influence of a (trait of a) molecular entity at the current time on those at future time points is assessed by mutual information, impulse response analysis, and path decomposition of the covariance. The presented methodology is applied to the omics data from the p53 signaling pathway during HPV-induced cellular transformation. All methodology is implemented in the ragt2ridges package, freely available from the Comprehensive R Archive Network.

Original languageEnglish
Pages (from-to)391-405
Number of pages15
JournalBiometrical Journal
Volume61
Issue number2
Early online date22 Aug 2018
DOIs
Publication statusPublished - Mar 2019

Fingerprint

Ridge
Network Model
Path Decomposition
Vector Autoregressive Model
Penalized Likelihood
Empirical Bayes
Signaling Pathways
Methodology
Regulatory Networks
Impulse Response
Thresholding
Cellular Networks
Mutual Information
Prior Knowledge
Network Topology
Exploitation
Penalty
Model
Network model
Estimate

Bibliographical note

Special Issue: ISCB38, Part I

Keywords

  • cervical cancer
  • constrained estimation
  • maximum likelihood
  • mutual information
  • path analysis
  • penalized estimation
  • time series analysis
  • vector autoregressive process

Cite this

Miok, Viktorian ; Wilting, Saskia M. ; van Wieringen, Wessel N. / Ridge estimation of network models from time-course omics data. In: Biometrical Journal. 2019 ; Vol. 61, No. 2. pp. 391-405.
@article{3cd4b4aaf658470b8e228dd0aff6e295,
title = "Ridge estimation of network models from time-course omics data",
abstract = "Time-course omics experiments enable the reconstruction of the dynamics of the cellular regulatory network. Here, we describe the means for this reconstruction and the downstream exploitation of the inferred network. It is assumed that one of the various vector-autoregressive models (VAR) models presented here serves as a reasonably accurate description of the time-course omics data. The models are estimated through ridge penalized likelihood maximization, accompanied by functionality for the determination of optimal penalty paramaters. Prior knowledge on the network topology is accommodated by the estimation procedures. Various routes that translate the fitted models into more tangible implications for the medical researcher are described. The network is inferred from the—nonsparse—ridge estimates through empirical Bayes probabilistic thresholding. The influence of a (trait of a) molecular entity at the current time on those at future time points is assessed by mutual information, impulse response analysis, and path decomposition of the covariance. The presented methodology is applied to the omics data from the p53 signaling pathway during HPV-induced cellular transformation. All methodology is implemented in the ragt2ridges package, freely available from the Comprehensive R Archive Network.",
keywords = "cervical cancer, constrained estimation, maximum likelihood, mutual information, path analysis, penalized estimation, time series analysis, vector autoregressive process",
author = "Viktorian Miok and Wilting, {Saskia M.} and {van Wieringen}, {Wessel N.}",
note = "Special Issue: ISCB38, Part I",
year = "2019",
month = "3",
doi = "10.1002/bimj.201700195",
language = "English",
volume = "61",
pages = "391--405",
journal = "Biometrical Journal",
issn = "0323-3847",
publisher = "Wiley-VCH Verlag",
number = "2",

}

Ridge estimation of network models from time-course omics data. / Miok, Viktorian; Wilting, Saskia M.; van Wieringen, Wessel N.

In: Biometrical Journal, Vol. 61, No. 2, 03.2019, p. 391-405.

Research output: Contribution to JournalArticleAcademicpeer-review

TY - JOUR

T1 - Ridge estimation of network models from time-course omics data

AU - Miok, Viktorian

AU - Wilting, Saskia M.

AU - van Wieringen, Wessel N.

N1 - Special Issue: ISCB38, Part I

PY - 2019/3

Y1 - 2019/3

N2 - Time-course omics experiments enable the reconstruction of the dynamics of the cellular regulatory network. Here, we describe the means for this reconstruction and the downstream exploitation of the inferred network. It is assumed that one of the various vector-autoregressive models (VAR) models presented here serves as a reasonably accurate description of the time-course omics data. The models are estimated through ridge penalized likelihood maximization, accompanied by functionality for the determination of optimal penalty paramaters. Prior knowledge on the network topology is accommodated by the estimation procedures. Various routes that translate the fitted models into more tangible implications for the medical researcher are described. The network is inferred from the—nonsparse—ridge estimates through empirical Bayes probabilistic thresholding. The influence of a (trait of a) molecular entity at the current time on those at future time points is assessed by mutual information, impulse response analysis, and path decomposition of the covariance. The presented methodology is applied to the omics data from the p53 signaling pathway during HPV-induced cellular transformation. All methodology is implemented in the ragt2ridges package, freely available from the Comprehensive R Archive Network.

AB - Time-course omics experiments enable the reconstruction of the dynamics of the cellular regulatory network. Here, we describe the means for this reconstruction and the downstream exploitation of the inferred network. It is assumed that one of the various vector-autoregressive models (VAR) models presented here serves as a reasonably accurate description of the time-course omics data. The models are estimated through ridge penalized likelihood maximization, accompanied by functionality for the determination of optimal penalty paramaters. Prior knowledge on the network topology is accommodated by the estimation procedures. Various routes that translate the fitted models into more tangible implications for the medical researcher are described. The network is inferred from the—nonsparse—ridge estimates through empirical Bayes probabilistic thresholding. The influence of a (trait of a) molecular entity at the current time on those at future time points is assessed by mutual information, impulse response analysis, and path decomposition of the covariance. The presented methodology is applied to the omics data from the p53 signaling pathway during HPV-induced cellular transformation. All methodology is implemented in the ragt2ridges package, freely available from the Comprehensive R Archive Network.

KW - cervical cancer

KW - constrained estimation

KW - maximum likelihood

KW - mutual information

KW - path analysis

KW - penalized estimation

KW - time series analysis

KW - vector autoregressive process

UR - http://www.scopus.com/inward/record.url?scp=85052394344&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85052394344&partnerID=8YFLogxK

U2 - 10.1002/bimj.201700195

DO - 10.1002/bimj.201700195

M3 - Article

VL - 61

SP - 391

EP - 405

JO - Biometrical Journal

JF - Biometrical Journal

SN - 0323-3847

IS - 2

ER -