Use of molecular docking and similarity models employing substances with biological activity against PPAR-delta receptor
Diabetes mellitus (DM) is a chronic disease characterized by a metabolic disorder arising from defects in insulin secretion and/or action, which results in hyperglycemia and causes a series of physiological complications. Considering that the disease has no cure and that the existing treatments have adverse effects that compromise the quality of life of patients, it is necessary to search for more effective treatments. The PPARs (peroxisome proliferator-activated receptor) controls the metabolic pathways of carbohydrates and lipids. A subclass of these receptors, the subtype PPARδ, regulates certain metabolic pathways so that substances that activate it can be used as drugs for type 2 DM. In this work, docking analyses were performed in order to understand the differences between the most and least active ligands selected at the target binding site. This technique showed that the most active ligands make hydrogen bonds with 3 of the major residues responsible for receptor activation, as well as a greater number of hydrophobic and van der Waals interactions.An exhaustive study was also made from the construction and validation of several chemical similarity models (CSMs) to understand which features were relevant for the activation of PPARδ and thus apply the best models in a future best models in a future virtual detrending protocol based on the structure of the ligand structure. We selected 2 CSMs that presented statistical parameter values such as MCC statistical parameter values, such as MCC, AUC and FNR, related to good predictive predictive ability. Furthermore, these models indicated the importance of the hydrogen bonding acceptor groups located in the "head" of the molecule (these groups are responsible for interacting with key residues for the activation of the receptor), corroborating data from the literature.