Building Kinetic Models with Complex Drug-Protein Interactions: application to the targeted inhibition of MAPK signaling in cancer

A key goal in the field of Quantitative Systems Pharmacology (QSP) is the construction of mechanistic models able to predict drug efficacy. A major challenge in building such models is the necessity to properly describe highly cooperative drug-protein and protein-protein interactions that govern the functioning of biochemical networks. In this seminar, I will show how Ordinary Differential Equations (ODEs) models comprising large numbers of drug-protein and protein-protein interactions can be efficiently built using rule-based modelling and energy-based descriptions of molecular cooperativity.The modelling framework I will present is based on an extension of the Python Systems Biology (PySB) toolbox to incorporate energy-based specifications supported by BioNetGen (eBNG). The resulting framework allows modelers to write large ODEs models as compact Python programs in which molecular cooperativity is specified as free energy contributions and detailed balance is satisfied by construction. As a case study, I will show that the framework allows the accurate description of high-order cooperativity interactions between components of the MAPK signaling pathway and targeted kinase inhibitors and that the inclusion of such interactions predicts clinically-relevant drug resistance mechanisms in skin and colorectal cancers.