Inferring network structure from interventional time-course experiments

TL;DR

This paper introduces a causal dynamic Bayesian network model that effectively integrates interventional data into time-course biological network inference, demonstrating improved accuracy on simulated and experimental data.

Contribution

It presents a novel causal variant of dynamic Bayesian networks that incorporates interventions, enhancing network inference from interventional time-course data.

Findings

Interventions must be properly modeled for accurate network inference.

The proposed model outperforms traditional methods on simulated data.

Empirical results validate the approach on experimental biological data.

Abstract

Graphical models are widely used to study biological networks. Interventions on network nodes are an important feature of many experimental designs for the study of biological networks. In this paper we put forward a causal variant of dynamic Bayesian networks (DBNs) for the purpose of modeling time-course data with interventions. The models inherit the simplicity and computational efficiency of DBNs but allow interventional data to be integrated into network inference. We show empirical results, on both simulated and experimental data, that demonstrate the need to appropriately handle interventions when interventions form part of the design.