L1-Penalized Censored Gaussian Graphical Model

TL;DR

This paper introduces an L1-penalized Gaussian graphical model tailored for censored data, addressing challenges in inferring genetic networks from high-dimensional, censored measurements, with demonstrated efficiency and practical application to gene expression data.

Contribution

It proposes a novel L1-penalized Gaussian graphical model for censored data and develops two EM-like algorithms for efficient inference, outperforming existing methods.

Findings

Algorithms are computationally efficient and effective with censored data.

The method outperforms competitors in simulation studies.

Applied successfully to gene expression data from microfluidic RT-qPCR.

Abstract

Graphical lasso is one of the most used estimators for inferring genetic networks. Despite its diffusion, there are several fields in applied research where the limits of detection of modern measurement technologies make the use of this estimator theoretically unfounded, even when the assumption of a multivariate Gaussian distribution is satisfied. Typical examples are data generated by polymerase chain reactions and flow cytometer. The combination of censoring and high-dimensionality make inference of the underlying genetic networks from these data very challenging. In this paper we propose an $\ell_1$-penalized Gaussian graphical model for censored data and derive two EM-like algorithms for inference. By an extensive simulation study, we evaluate the computational efficiency of the proposed algorithms and show that our proposal overcomes existing competitors when censored data are available. We apply the proposed method to gene expression data coming from microfluidic RT-qPCR technology in order to make inference on the regulatory mechanisms of blood development.