A Comparative Study Of Different Virtual Screening Tools In View Of Structure

With the rapid development of computational technology,and the dramatically increasing need for medicine of world wide population,drug discovery has become a hot issue in the past 60 years.Computer aided drug design(CADD)has played an ever more increasing role in the drug discovery process.Traditional methods for drug screening produce heavy load for research institutions and pharmaceutical companies,and so in silico virtual screening is frequently used to optimize high-throughput screening experiments.As a result,the cost of drug discovery can be driven down to a lower level.Distributed computing refers to the use of distributed systems to solve computational problems,in which the basic components located on communicate and coordinate their actions by passing messages in distributed computers.A task is divided into many sub jobs,each of which is solved by one or more processors.For each distributed job there is a driver host that is responsible for dividing up the work,submitting the sub jobs,and integrating the results.Depending on the program,the driver can run on the local host or on the remote host.Hardware and software debugging are the key steps in the construction of a high throughput virtual screening pipeline,in which the compatibility of computation nodes,data synchronization,result integration and convention scripts are all important factors.The suitable selection of the scoring function is also an issue we met during the process of method validation.We chose Aurora-A kinase and lactamase enzyme HDM-1 which have several known inhibitors as targets for comparative analysis of a variety of screening methods to justify distributed computing based on virtual screening methods.we tried multiple sorts of possible solutions and built the virtual screening pipeline.Aurora A is a member of the regulations of cell proliferation and cancer development associated Aurora kinase family.HDM-1 actamase is a class of enzymes secreted by resistant bacteria against lactam antibioticsa.The database with more than 690000 moleculars was tested to obtain the screening speed and capacity.The drug-like compound library which contains 6.6 million molecules was screen with our virtual screening system,421 compounds of 13 clusters were obtained as a result of virtual screening.After visual inspection,12 compounds were chosen for purchasing and synthesis,after that biological activity testing and structural optimization steps were carried on,1 compound with a higher inhibitory activity of NDM-1 was obtained finally.