QSAR AND MOLECULAR DOCKING STUDY OF GONADOTROPIN-RELEASING HORMONE RECEPTOR INHIBITORS

Abstract

BackgroundA case study of activities of pyrimidine-based molecules to the gonadotropin-releasing hormone receptor is used to develop a QSAR model. The chemical dataset collected as CHEMBL1855 were curated in line with published works of Prof. Gramatica. The parameters for QSAR model validation and ability to predict activities of future chemicals (internal robustness, exclusion of chance correlation, external predictivity, applicability domain) are presented in the work, while the molecular descriptors used for the model were calculated using the free PaDEL-Descriptor software. Other important information regarding the model was done with the aid of the QSARINS. This work uses the Setubal principle as a guideline in hypothesizing a well outlined step in designing novel more active compounds using a ligand based-design approach. We highlighted a list of things to be considered when selecting a lead compound, after which the hypothesis was tested and a series of novel compounds were manually designed. ResultThe QSAR models initially developed was used to predict the activities of the new compounds before subjecting the novel compounds a molecular docking study. The external validation parameter R2ext = 0.5328, while the internal cross validation parameter for the model was found to be Q2loo= 0.5316, this shows that the model could be used to predict the activitie of other inhibitor within its chemical space. The model was used to design several compound with better inhibitory potential than their lead or parent structures. One of such compounds B8, was ported to have a calculated binding energy of -30.442 kcal/mol, which was significantly lower than the binding energy of the parent compound (LeadB) which is -21.373 kcal/mol. The calculated pKI value of B8 and other designed novel compounds were predicted with the developed model was found to be better than their parent compounds used in their design. ConclusionThe predicted pKI values from the QSAR predictions and the Molecular docking studies of the novel molecules where compared with their parent Lead (molecules) and were found to be significantly better, the result showcases the merit in applying the proposed technique in the design of future drug candidates.