Abstract
The elucidation of gene regulatory networks is one of the major challenges of systems biology. Measurements about genes that are exploited by network inference methods are typically available either in the form of steady-state expression vectors or time series expression data. In our previous work, we proposed the GENIE3 method that exploits variable importance scores derived from Random forests to identify the regulators of each target gene. This method provided state-of-the-art performance on several benchmark datasets, but it could however not specifically be applied to time series expression data. We propose here an adaptation of the GENIE3 method, called dynamical GENIE3 (dynGENIE3), for handling both time series and steady-state expression data. The proposed method is evaluated extensively on the artificial DREAM4 benchmarks and on three real time series expression datasets. Although dynGENIE3 does not systematically yield the best performance on each and every network, it is competitive with diverse methods from the literature, while preserving the main advantages of GENIE3 in terms of scalability.
Highlights
Gene regulatory networks (GRNs) define the ensemble of interactions among genes that govern their expression
The proposed variant for time series data, called dynGENIE3, is based on a semi-parametric model, in which the temporal evolution of each gene expression is described by an ordinary differential equation (ODE) and the transcription function in each ODE is learned in the form of a non-parametric Random forest model
We first evaluated the performances of GENIE3 and dynGENIE3 on the simulated data of the DREAM4 In Silico Network challenge
Summary
Gene regulatory networks (GRNs) define the ensemble of interactions among genes that govern their expression. Steady-state expression data are plethoric for many organisms They offer limited information regarding the dynamics of gene regulation, which limits the performance of network inference methods when they only exploit such data. We proposed GENIE3, a model-free method that infers networks from steady-state expression data[6] This method exploits variable importance scores derived from Random forests[10] to identify the regulators of each target gene. The proposed variant for time series data, called dynGENIE3 (for dynamical GENIE3), is based on a semi-parametric model, in which the temporal evolution of each gene expression is described by an ordinary differential equation (ODE) and the transcription function in each ODE is learned in the form of a non-parametric Random forest model. The regulators of each target gene are identified from the variable importance scores derived from the corresponding Random forest model
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
