Ensemble inference and inferability of gene regulatory networks.

S M Minhaz Ud-Dean,Rudiyanto Gunawan,Panayiotis V Benos

doi:10.1371/journal.pone.0103812

S M Minhaz Ud-Dean, Rudiyanto Gunawan + Show 1 more

Open Access

https://doi.org/10.1371/journal.pone.0103812

Copy DOI

Abstract

The inference of gene regulatory network (GRN) from gene expression data is an unsolved problem of great importance. This inference has been stated, though not proven, to be underdetermined implying that there could be many equivalent (indistinguishable) solutions. Motivated by this fundamental limitation, we have developed new framework and algorithm, called TRaCE, for the ensemble inference of GRNs. The ensemble corresponds to the inherent uncertainty associated with discriminating direct and indirect gene regulations from steady-state data of gene knock-out (KO) experiments. We applied TRaCE to analyze the inferability of random GRNs and the GRNs of E. coli and yeast from single- and double-gene KO experiments. The results showed that, with the exception of networks with very few edges, GRNs are typically not inferable even when the data are ideal (unbiased and noise-free). Finally, we compared the performance of TRaCE with top performing methods of DREAM4 in silico network inference challenge.

Highlights

The discovery and analysis of biological networks have important applications, from finding treatment of diseases to engineering of microbes for the production of drugs and biofuels [1,2,3,4]
We introduce new framework and algorithms, called Transitive Reduction and Closure Ensemble (TRaCE), for the ensemble inference of gene regulatory network (GRN)
We investigated the inferability of random GRNs of orders n~10 and n~100 genes

Summary

Introduction

The discovery and analysis of biological networks have important applications, from finding treatment of diseases to engineering of microbes for the production of drugs and biofuels [1,2,3,4]. The Dialogue for Reverse Engineering Assessments and Methods (DREAM) project is a community-wide effort initiated to fulfill the need for a rigorous and fair comparison of the strengths and weaknesses of methods for the reverse engineering of biological networks from data. To this end, challenges involving the inference of cellular networks are organized on a yearly basis (http://www.the-dream-project.org/challenges). The inference of GRN has become a major focus of several DREAM challenges The outcomes of these challenges indicate that the state-of-the-art algorithms for GRN inference are unable to provide accurate and reliable network predictions, even when large expression datasets are available and the number of genes is small (10–100 genes) [8,9,10]. A crowd-sourcing strategy that combines the predictions of different inference methods has been shown to be more reliable than any individual method [10]

Methods

Results

Discussion

Conclusion