Dynamic Bayesian Network Learning to Infer Sparse Models From Time Series Gene Expression Data

Abstract

One of the key challenges in systems biology is to derive gene regulatory networks (GRNs) from complex high-dimensional sparse data. Bayesian networks (BNs) and dynamic Bayesian networks (DBNs) have been widely applied to infer GRNs from gene expression data. GRNs are typically sparse but traditional approaches of BN structure learning to elucidate GRNs often produce many spurious (false positive) edges. We present two new BN scoring functions, which are extensions to the Bayesian Information Criterion (BIC) score, with additional penalty terms and use them in conjunction with DBN structure search methods to find a graph structure that maximises the proposed scores. Our BN scoring functions offer better solutions for inferring networks with fewer spurious edges compared to the BIC score. The proposed methods are evaluated extensively on auto regressive and DREAM4 benchmarks. We found that they significantly improve the precision of the learned graphs, relative to the BIC score. The proposed methods are also evaluated on three real time series gene expression datasets. The results demonstrate that our algorithms are able to learn sparse graphs from high-dimensional time series data. The implementation of these algorithms is open source and is available in form of an R package on GitHub at https://github.com/HamdaBinteAjmal/DBN4GRN, along with the documentation and tutorials.

Gat2Get: A Novel Approach to Infer Gene Regulatory Network from Gene Activity using Dynamic Bayesian Network learning

“…BN as a classifier has the capacity to forecast the likelihood of class members, i.e., the likelihood that a specified set of data is part of a specific class. Ajmal and Madden in [1] discuss the Bayesian Network (BN) as a straightforward diagram where every node is a data member, and each edge denotes a probabilistic relationship as (see Figure 3) [11].…”

“…Gene Regulatory Network (GRN) is known as a group of genes that indirectly regulate the activity rates of each other through their protein or RNA products [1]. Discovering and understanding this regulation processes that underlie the vital operations of human disease is one of the major goals in bioinformatics.…”

“…A significant new tool for measuring the gene expression profile of known coding sequences in a particular tissue and time point is microarray innovation. The recent advances of this technology help to propose many approaches to infer the gene networks from these experimental data [1].…”

“…Data mining is one of many techniques that try to do that. In order to predict regulatory networks from gene expression patterns, a variety of algorithms, including Bayesian network algorithms [1], neural networks [2], and Boolean networks [2], are utilized. In general, there are two approaches to build and model GRN, the first approach is based on finding the master regulators set of GRN and obtaining GRN.…”

Discovering gene regulatory networks of multiple phenotypic groups using dynamic Bayesian networks

“…DBN models were used to learn biological networks [ 35 ], including GRNs [ 2 , 6 , 15 , 30 , 59 , 64 ] and multi-omics networks [ 48 ]. Learning DBN structures from data is computationally challenging because the number of possible network topologies grows exponentially with the number of nodes.…”