Structure-Based Virtal Screening

Question: Discuss the structure-Based virtal screening and scoring functions for protein-Ligand docking. Answer: Structure-Based Virtal Screening With increase in number of crystal structures of protein, interest to use that detailed knowledge of structure for data analysis, library design and compound acquisition. Design method based on structure has been developed over number of year. For thorough analysis of the small number of the molecules number of activities was conducted toward the design method based on structure. For higher and throughout analysis of structure based methods number of factors contributed. First, unparalleled amount of hard-working computing at less cost has provided by computers hardwares with higher performance (usually based on operating system that is linux) to the researchers. Second, for the molecular docking new algorithms has been developed by expending significant efforts. Finally, analysis tools for output result analysis of calculations empower scientist to work more effectively with the huge number of quantities of generated data. Objective of experiment docking is to conclude the three dimensional structure-formed while single or number of molecules constructs an intermolecular complex. For ligand of protein docking number of techniques has been suggested [Dixon and Blaney 1993; Totrov and Abagyan 2001; Halperin et al 2002; Taylor et al 2002]. For problems related to docking there are mainly two components. First, there should be mechanism to analyze space of desirable proteinligand geometry called pose. Next, it should be able to rank or score that pose for identifying mode of binding for every compound and for assigning priority order to the molecules. Main difficulty is in the part in ligand of protein docking because number of degree of freedom involved in it. In the rotation and translation of single molecule with respect to other molecules, there are six degrees of freedoms involved. In addition there are conformational degrees of freedoms of protein and ligand. For determination of geometry of free energy of binding and ligand of protein, solvent also plays important role but thats sometimes ignored. A master of computational-chemist may capable to analyze the binding mode of ligand by use of interactive molecular graphic if she or he has good knowledge of mode of binding (e.g if there are X ray structures of close analogues available). So, manual docking is impractical for a large number of molecules and can be very difficult while dealing with the novel ligands structures. According to degree of freedom they consider it becomes convenient to categorize the different docking algorithm. Earliest algorithm just considers the rotational and translational degree of freedom of ligand and protein, by treating both bodies as rigid. At present mostly using algorithm enables ligands to fully examine their conformational degree of freedom; few program also permit very definite conformational flexibility to the protein (e.g. by permitting to rotate without obstruction to side chain proton of threonine, lysine, serine and trysin residue). Some developed algorithm tried with taking greater protein flexibility into account, but they are neither in wide spread use nor been evaluated extensively. [Shoichet et al 2002; Carlson. 2002] Algorithm DOCK develops by, Kuntz with co-workers is one of most advance algorithm in docking of ligand of protein [Desjarlais et al 1988; Kuntz et al 1994; Kuntz 1992; Kuntz et al 1982]. Latest version of algorithm DOCK considered only docking of body those are rigid and design to determine molecule with high degrees of shapes complementarities to binding site of protein. Initial stages of method of DOCK involve creation of Negative image of binding sites. Sequence of overlapping sphere of differing radius, explore from protein molecular surface involves in that negative image. Recent algorithm those take conformational degree of freedom of ligand into consideration can easily classified according to the method they explore conformational space. Easiest approach to make and ensemble of conformation of ligand by using convenient search algorithms of conformational and then by using a rigid body algorithm, dock the each of them. That approach may seem to ineffective approach, but yet numbers of effective method are worked on basis of this approach; these methods involved a variation of filter and approximation to quickly determine the conformation of the interests [FRED, Glide]. The purpose of doing conformational analysis by using ligand those are isolated may be questionable but a lot of analysis have shown that conformation of ligand noticed in ligand of protein complexes are usually has similar geometry to reduce energy conformation of ligand those are isolated [Bostrom et al 1998]. Alternative methods show the conformational and orientational degree of f reedom simultaneous. From these methods mostly fall in three categories: Genetic algorithm, Monte Carlo algorithm and incremental construction approach. For protein ligand docking easiest Monte Carlo algorithm are nearly relevant to those working for the conformational analysis [Olsen and Goodsell 1990]. To carry out ligand of protein docking, we can use Evolutionary and Genetic algorithm [Gehlhaar et al 1995; Oshiro et al 1995; Jones et al 1995b Judson et al 1994;]. In series of periods within binding site, Incremental construction method constructs conformation of ligand. [Kuntz and Leach 1990; Rarey et al 1996; Welch et al 1996] Scoring Functions for Protein-Ligand Docking In experiment of virtual screening based on structure, its generally useful for making discrepancy in between scoring and docking. Purpose of docking is to determine the location thats nearest in the geometry to structure checked that is X-ray because docking involves prediction of mode of binding of molecules those is individual. To determine the performance of docking programs no. of analysis has been conducted by using derived data from Protein Data Bank and from these analysis its found that in 70%+ cases docking programs correctly predict binding geometry.[Kramer et al 1999; Jones et al 1997; Nissink et al 2002]. Results produced from typical program of docking are displayed in Fig. 8-4.It is important for virtual screening to be capable to rank or score ligands by using functions related to association of free energy of ligand and protein for formation of intermolecular complex