Predicting Moonlighting Proteins to Prevent Drug Side Effects

Undesired side effects, which threaten both patients and pharmaceutical companies, sometimes occur when a drug designed for a particular purpose affects proteins carrying unexpected functions. Benoît Robisson aims to identify such moonlighting proteins.
Research on side-effects of drugs became an important subject of studies in the pharmaceutical field. Side-effects may be encountered when the chosen drug target is a moonlighting protein (i.e. protein with several unrelated functions). Therefore, this kind of proteins remains widely unknown, and there is a need to develop a method to identify them. We propose a large-scale and without a priori method based on the partitioning of protein-protein interactions networks. Proteins interact to perform there functions, and the partitioning group in a same class proteins involved in a same biological process. Then, proteins find at the intersection of several classes are good moonlighting proteins candidates.