Prioritization of Candidate Genes Through Boolean Networks

The in silico detection of master regulator genes is a popular attempt at speeding up drug development. These genes might be directly related to the onset of the disease, or may act on one pathway which counteracts the associated symptoms. Then, one could perhaps screen drugs to select chemical compounds targeting these genes. In prior works, the detection of these candidates was performed through the identification of the regulatory interactions between genes of interest for the disease. Indeed, system biology approaches have proven a useful tool to integrate transcriptomic data and predict transcriptional profiles under gene perturbations. However, for rare or tropical neglected diseases, building such a regulatory model can become a tedious and time-consuming task. In this work, we show how to build, in a reproducible and transparent fashion, a gene regulatory network using publicly available data. Then, we describe a method to identify master regulatory genes, which have an impact on the dynamics of the gene regulation in a specific disease-related transcriptional context. We showed that our novel method for the identification of master regulatory genes was consistent with network controllability measures, while targeting genes that were significantly enriched for epilepsy-related terms. Our pipeline allows for systematic and transparent Boolean network synthesis, and identification of master re-gulators, which might help tackle the issue of rare or tropical neglected diseases.