Yapı odaklı ilaç tasarımı

The aim of this study is to target two possible pathways of cancer which are (i) p65p50 protein heterodimer from NF-?B protein family and (ii) Fgfr2 from fibroblastic growth hormone receptors. A structure-based screening approach, based on a data set of 236 natural products for each protein is adopted. Docking analysis is performed for structure-based screening. Three scoring functions are used for docking. Docking results of both proteins and their ranking are compared using known scoring functions. It is shown that scoring functions have a tendency to rank specific structural groups selectively. AutoDock, which is based on a semi-empirical function, performs the rankings according to the properties of the ligands in the data set. Hence, AutoDock ranks via selecting a specific group from the data set. The two other scoring functions, Gold Score and Chem Score that come with the software GOLD (Genetic Optimization for Ligand Docking) show that specific groups of molecules are ranked selectively according to the properties of both the system and the ligand. Variation between selectivity of different scoring functions is based on the main equations used within each scoring function. On the other hand, as also reported in recent studies, although the ranking of a docked protein-ligand pair depends on the scoring function used, the orientation and interaction modes of docking performed with different scoring functions are similar.