Gene network reconstruction using global-local shrinkage priors

G.G.R. Leday, M.C.M. de Gunst, Gino B. Kpogbezan, A.W. van der Vaart, W.N. van Wieringen, M.A. van de Wiel

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

Reconstructing a gene network from high-throughput molecular data is an important but challenging task, as the number of parameters to estimate easily is much larger than the sample size. A conventional remedy is to regularize or penalize the model likelihood. In network models, this is often done locally in the neighborhood of each node or gene. However, estimation of the many regularization parameters is often difficult and can result in large statistical uncertainties. In this paper we propose to combine local regularization with global shrinkage of the regularization parameters to borrow strength between genes and improve inference. We employ a simple Bayesian model with nonsparse, conjugate priors to facilitate the use of fast variational approximations to posteriors. We discuss empirical Bayes estimation of hyperparameters of the priors, and propose a novel approach to rank-based posterior thresholding. Using extensive model- and data-based simulations, we demonstrate that the proposed inference strategy outperforms popular (sparse) methods, yields more stable edges, and is more reproducible. The proposed method, termed ShrinkNet, is then applied to Glioblastoma to investigate the interactions between genes associated with patient survival.
Original languageEnglish
Pages (from-to)41-68
JournalThe Annals of Applied Statistics
Volume11
Issue number1
DOIs
Publication statusPublished - 2017

Fingerprint

Gene Networks
Shrinkage
Genes
Regularization Parameter
Gene
Empirical Bayes Estimation
Variational Approximation
Conjugate prior
Hyperparameters
Bayesian Model
Thresholding
Network Model
High Throughput
Likelihood
Regularization
Sample Size
Uncertainty
Throughput
Vertex of a graph
Interaction

Cite this

Leday, G.G.R. ; de Gunst, M.C.M. ; Kpogbezan, Gino B. ; van der Vaart, A.W. ; van Wieringen, W.N. ; van de Wiel, M.A. / Gene network reconstruction using global-local shrinkage priors. In: The Annals of Applied Statistics. 2017 ; Vol. 11, No. 1. pp. 41-68.
@article{60cd2f5d564949388933ca628d90883f,
title = "Gene network reconstruction using global-local shrinkage priors",
abstract = "Reconstructing a gene network from high-throughput molecular data is an important but challenging task, as the number of parameters to estimate easily is much larger than the sample size. A conventional remedy is to regularize or penalize the model likelihood. In network models, this is often done locally in the neighborhood of each node or gene. However, estimation of the many regularization parameters is often difficult and can result in large statistical uncertainties. In this paper we propose to combine local regularization with global shrinkage of the regularization parameters to borrow strength between genes and improve inference. We employ a simple Bayesian model with nonsparse, conjugate priors to facilitate the use of fast variational approximations to posteriors. We discuss empirical Bayes estimation of hyperparameters of the priors, and propose a novel approach to rank-based posterior thresholding. Using extensive model- and data-based simulations, we demonstrate that the proposed inference strategy outperforms popular (sparse) methods, yields more stable edges, and is more reproducible. The proposed method, termed ShrinkNet, is then applied to Glioblastoma to investigate the interactions between genes associated with patient survival.",
author = "G.G.R. Leday and {de Gunst}, M.C.M. and Kpogbezan, {Gino B.} and {van der Vaart}, A.W. and {van Wieringen}, W.N. and {van de Wiel}, M.A.",
year = "2017",
doi = "10.1214/16-AOAS990",
language = "English",
volume = "11",
pages = "41--68",
journal = "The Annals of Applied Statistics",
issn = "1932-6157",
publisher = "Institute of Mathematical Statistics",
number = "1",

}

Gene network reconstruction using global-local shrinkage priors. / Leday, G.G.R.; de Gunst, M.C.M.; Kpogbezan, Gino B.; van der Vaart, A.W.; van Wieringen, W.N.; van de Wiel, M.A.

In: The Annals of Applied Statistics, Vol. 11, No. 1, 2017, p. 41-68.

Research output: Contribution to JournalArticleAcademicpeer-review

TY - JOUR

T1 - Gene network reconstruction using global-local shrinkage priors

AU - Leday, G.G.R.

AU - de Gunst, M.C.M.

AU - Kpogbezan, Gino B.

AU - van der Vaart, A.W.

AU - van Wieringen, W.N.

AU - van de Wiel, M.A.

PY - 2017

Y1 - 2017

N2 - Reconstructing a gene network from high-throughput molecular data is an important but challenging task, as the number of parameters to estimate easily is much larger than the sample size. A conventional remedy is to regularize or penalize the model likelihood. In network models, this is often done locally in the neighborhood of each node or gene. However, estimation of the many regularization parameters is often difficult and can result in large statistical uncertainties. In this paper we propose to combine local regularization with global shrinkage of the regularization parameters to borrow strength between genes and improve inference. We employ a simple Bayesian model with nonsparse, conjugate priors to facilitate the use of fast variational approximations to posteriors. We discuss empirical Bayes estimation of hyperparameters of the priors, and propose a novel approach to rank-based posterior thresholding. Using extensive model- and data-based simulations, we demonstrate that the proposed inference strategy outperforms popular (sparse) methods, yields more stable edges, and is more reproducible. The proposed method, termed ShrinkNet, is then applied to Glioblastoma to investigate the interactions between genes associated with patient survival.

AB - Reconstructing a gene network from high-throughput molecular data is an important but challenging task, as the number of parameters to estimate easily is much larger than the sample size. A conventional remedy is to regularize or penalize the model likelihood. In network models, this is often done locally in the neighborhood of each node or gene. However, estimation of the many regularization parameters is often difficult and can result in large statistical uncertainties. In this paper we propose to combine local regularization with global shrinkage of the regularization parameters to borrow strength between genes and improve inference. We employ a simple Bayesian model with nonsparse, conjugate priors to facilitate the use of fast variational approximations to posteriors. We discuss empirical Bayes estimation of hyperparameters of the priors, and propose a novel approach to rank-based posterior thresholding. Using extensive model- and data-based simulations, we demonstrate that the proposed inference strategy outperforms popular (sparse) methods, yields more stable edges, and is more reproducible. The proposed method, termed ShrinkNet, is then applied to Glioblastoma to investigate the interactions between genes associated with patient survival.

U2 - 10.1214/16-AOAS990

DO - 10.1214/16-AOAS990

M3 - Article

VL - 11

SP - 41

EP - 68

JO - The Annals of Applied Statistics

JF - The Annals of Applied Statistics

SN - 1932-6157

IS - 1

ER -