RACIPE: a computational tool for modeling gene regulatory circuits using randomization

Bin Huang,Herbert Levine,Mingyang Lu,Jingchen Feng,Dongya Jia,José N Onuchic

doi:10.1186/s12918-018-0594-6

Bin Huang, Herbert Levine + Show 4 more

Open Access

https://doi.org/10.1186/s12918-018-0594-6

Copy DOI

Abstract

BackgroundOne of the major challenges in traditional mathematical modeling of gene regulatory circuits is the insufficient knowledge of kinetic parameters. These parameters are often inferred from existing experimental data and/or educated guesses, which can be time-consuming and error-prone, especially for large networks.ResultsWe present a user-friendly computational tool for the community to use our newly developed method named random circuit perturbation (RACIPE), to explore the robust dynamical features of gene regulatory circuits without the requirement of detailed kinetic parameters. Taking the network topology as the only input, RACIPE generates an ensemble of circuit models with distinct randomized parameters and uniquely identifies robust dynamical properties by statistical analysis. Here, we discuss the implementation of the software and the statistical analysis methods of RACIPE-generated data to identify robust gene expression patterns and the functions of genes and regulatory links. Finally, we apply the tool on coupled toggle-switch circuits and a published circuit of B-lymphopoiesis.ConclusionsWe expect our new computational tool to contribute to a more comprehensive and unbiased understanding of mechanisms underlying gene regulatory networks. RACIPE is a free open source software distributed under (Apache 2.0) license and can be downloaded from GitHub (https://github.com/simonhb1990/RACIPE-1.0).

Highlights

One of the major challenges in traditional mathematical modeling of gene regulatory circuits is the insufficient knowledge of kinetic parameters
Time cost of simulations To evaluate the performance of the tool with different choices of simulation parameters, we test the tool on two types of coupled toggle-switch (CTS) circuits (Fig. 1b, see Additional file 1: SI section “Results” for mathematical models)
The total time to simulate all random circuit perturbation (RACIPE) models depends on other factors, which will be discussed

Summary

Introduction

One of the major challenges in traditional mathematical modeling of gene regulatory circuits is the insufficient knowledge of kinetic parameters. To model the dynamics of GRNs, different computational algorithms have been developed [12], such as ordinary differential equations (ODEs)-based models [13], Boolean network models [14, 15], Bayesian network models [16], It is believed that there is a core gene regulatory circuit underlying a GRN which functions as a decision-making module for one specific biological process [23, 24]. Identification of such core gene circuits can largely reduce the complexity of network modeling. Its operation is usually regulated by other genes and signaling pathways (“peripheral factors”) that interact with the core

Results

Discussion

Conclusion