Network Inference for Drug Discovery

Both in the preclinical and clinical stages, the drug discovery process is difficult and expensive. The pharmaceutical business is shifting away from a pathology-based approach and towards symptomatic alleviation. The delivery of medications whose targets are implicated in the disease’s causative processes should be made easier with a greater understanding of pathophysiology. The topology and the causal structure of the interactions between the medications and their targets are inferred from high-throughput experimental data using computational approaches in biological network inference. Therefore, biological network inference provides an expedited method for identifying the biochemical networks involved in drug metabolism and mechanisms of action as well as the gene and protein networks responsible for a disease. Pharmaceutical and biotechnology companies are increasingly using the generated high-level networks in drug development as a starting point for more intricate mechanistic models. In this chapter, we discuss and compare existing network inference computational tools and their potential to support the drug discovery process. In order to give the reader advice on how to select the optimal way based on the topic to be addressed, we highlight the applicability framework of several approaches.