A modified newton's method for inverse problem of Probabilistic Boolean Networks with gene perturbations

Abstract

Modeling genetic regulatory networks is an important research issue in systems biology. Many mathematical models have been proposed, and among these models, Boolean Network (BN) and its extension Probabilistic Boolean Network (PBN) are popular. In this paper we consider the problem constructing PBNs with gene perturbations. We propose a modified Newton's method to get the gene perturbation probability of the captured problem. Numerical experiments are given to demonstrate both effectiveness and efficiency of our proposed method.