Regularized Estimation of Large Scale Gene Regulatory Networks
نویسندگان
چکیده
When dealing with graphical Gaussian models for gene regulatory networks, the major problem is to compute the matrix of partial correlations. Based on the close connection between partial correlations and least squares regression, we suggest estimation of high-dimensional gene networks in terms of partial least squares (PLS) regression and the adaptive Lasso, respectively. In a simulation study, we compare the performance of the proposed methods in terms of their ability to estimate partial correlations and to derive the underlying network structure.
منابع مشابه
Efficient estimation of covariances and dependencies in high-dimensional gene expression data
The learning of dependencies in microarray data is challenging. Here, we will give a review of estimation methods based on Stein-type shrinkage. At their core lies a regularized estimation of the covariance matrix of the data. Subsequently, genetic networks from both, static and time-series data, can be inferred. The algorithms described exhibit a high accuracy, are computationally efficient, a...
متن کاملEstimation of nonlinear gene regulatory networks via L1 regularized NVAR from time series gene expression data.
Recently, nonlinear vector autoregressive (NVAR) model based on Granger causality was proposed to infer nonlinear gene regulatory networks from time series gene expression data. Since NVAR requires a large number of parameters due to the basis expansion, the length of time series microarray data is insufficient for accurate parameter estimation and we need to limit the size of the gene set stro...
متن کاملRegularized EM algorithm for sparse parameter estimation in nonlinear dynamic systems with application to gene regulatory network inference
Parameter estimation in dynamic systems finds applications in various disciplines, including system biology. The well-known expectation-maximization (EM) algorithm is a popular method and has been widely used to solve system identification and parameter estimation problems. However, the conventional EM algorithm cannot exploit the sparsity. On the other hand, in gene regulatory network inferenc...
متن کاملH∞ Sampled-Data Controller Design for Stochastic Genetic Regulatory Networks
Artificially regulating gene expression is an important step in developing new treatment for system-level disease such as cancer. In this paper, we propose a method to regulate gene expression based on sampled-data measurements of gene products concentrations. Inherent noisy behaviour of Gene regulatory networks are modeled with stochastic nonlinear differential equation. To synthesize feed...
متن کاملAn empirical Bayes approach to inferring large-scale gene association networks
MOTIVATION Genetic networks are often described statistically using graphical models (e.g. Bayesian networks). However, inferring the network structure offers a serious challenge in microarray analysis where the sample size is small compared to the number of considered genes. This renders many standard algorithms for graphical models inapplicable, and inferring genetic networks an 'ill-posed' i...
متن کامل