<p>In silico ligand docking is a versatile and common technique when predicting ligands and inhibitors for protein binding sites. The various docking programmes aim to calculate binding energies and to predict interactions, thus identifying potential ligands.The currently available programmes lack satisfying means by which to account for structural water molecules which can either mediate protein-ligand contacts or be displaced upon ligand binding. The present project aims to generate data to facilitate the global work of developing scoring functions in docking programmes to account for structural water molecules contribution to ligand binding to fill the said void. This is done by validating the performance of docking using a simple model system (cytochrome C peroxidase (CCP) W191G) containing four well ordered, deeply buried structural water molecules which are known to either interact with a ligand or to be displaced upon ligand binding.Known ligands were docked into eight (crystallographically determined) receptor set-ups comprising the receptor and no, one or two of the water molecules. The performance was validated by comparison of the binding modes of the docked ligands and the crystal structures, comparison of docking scores of the ligands in the different set-ups, enrichment of the ligands from a database of decoys and finally by predicting new ligands from the decoy database. In addition a high resolution crystal structure of CCP W191G in complex with 3-aminopyridine (3AP) was determined in order to resolve ambiguities in the binding mode of this ligand.</p>
| Identifer | oai:union.ndltd.org:UPSALLA/oai:DiVA.org:liu-19525 |
| Date | January 2009 |
| Creators | Wahlström, Rickard |
| Publisher | Department of Physics, Chemistry and Biology |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, text |
Page generated in 0.0025 seconds