Fibrates are peroxisome proliferator-activated alpha receptor (PPARα) activators derived from fibric acid and are the most clinically used therapeutics in the treatment of hypertriglyceridemia. Recently, we reported a computational approach for the investigation of the binding properties of fibrates, characterized by similar carboxylic heads but differing in the size and orientation of the hydrophobic portion. This procedure is based on a combination of standard docking and molecular mechanics approaches to better describe the adaptation of the protein target to the bound ligand. The application of our approach to a set of 23 fibrates and the use of an effective regression procedure, allowed the development of predictive models of the PPARα agonism. The obtained models are characterized by good performances realizing a fair trade-off between accuracy and computational costs. The best model is more specialized in the ranking of fibrate agonists whose binding is mainly controlled by steric rather than by electronic modulation. Here, we describe in details the application of this computational procedure for the prediction of PPARα agonism of fibrate ligands.