Inference of Gene Regulatory Networks Via Multiple Data Sources and a Recommendation Method

Gene regulatory networks (GRNs) are composed of biological components, including genes, proteins and metabolites, and their interactions. In general, computational methods are used to infer the connections among these components. However, computational methods should take into account the general features of the GRNs, which are sparseness, scale-free topology, modularity and structure of the inferred networks. In this work, observing the common aspects between recommendation systems and GRNs, we decided to map the GRNs inspiring problem into a recommendation problem and then used a known recommendation method to predict gene relationships based on multiple data sources, e.g., which molecules regulate others. The method we used is based on Pareto dominance and collaborative filtering. For the experiments, we used a combination of two datasets, namely microarray data and transcription factor (TF) binding data. The reported results show that using information from multiple sources improves the performance. Also, we observed that employing an approach from the recommendation systems domain revealed interesting results and good performance