Mechanistic Modeling of Genome Scale Molecular Interaction Networks

Physiologically-based Pharmacokinetic (PBPK) modeling has become the industry standard for predicting drug-drug interactions, formulation effects, and pharmacokinetics in human populations. As a bottom-up, literature-based, mechanistic computer simulation approach, it shares general methodology with the computational systems biology, where molecular biology knowledge is assimilated into molecular network models. In particular, reconstruction of Genome Scale Metabolic Networks (GSMNs) has led to mechanistic models incorporating a whole set of metabolic enzymes expressed in human tissues. Moreover, dynamic models of the expression of key drug metabolism enzymes are available.

Currently, PBPK models account for about 20 genes involved in drug metabolism. Integrating PBPK with GSMN and gene regulation models can extend the scope of mechanistic pharmacokinetic modeling to thousands of genes as well as the complex interactions of drug metabolism enzymes with endogenous metabolites such as cortisol. In this webinar, Dr. Andrzej Kierzek demonstrated how integrating human hepatocyte GSMN, gene regulation of CYP3A4, and a basic PBPK model can identify genes influencing the production of toxic metabolites and simulate kinetics of this metabolite as a function of cortisol and Pregnane X receptor (PXR) ligands.