Abstract:
It is well known that the costs associated with drug discovery are extremely high due the use of expensive in vitro methods as well as the high failure rate of drugs during clinical testing. In order to effectively fight the war against diseases such as malaria a far less expensive approach is required. Increasing the amount of in silico</i. work would decrease the amount of experiments that have to be done and so doing reduce the costs. In silico methods have the ability to predict major limiting factors in drug discovery resulting in only high quality drug molecules being subjected to clinical trials, thus increasing the probability of success. In this study a web-based chemoinformatics workspace aimed at research biologists was developed inside the FunGIMS application framework. This particular web-based chemoinformatics application was developed in the context of the FunGIMS suite of tools, specifically providing biological users with some of the most common chemoinformatics functions such as (1) the representation of chemical compounds, (2) chemical data, (3) databases and data sources, (4) structure search methods, (5) methods for calculating physical and chemical data (6) calculation of structure descriptors. The chemoinformatics module is supported by a modified version of the CHEBI database and the integration of OpenBabel and Frowns made it possible to perform analysis on molecules by means of SMILES-structures. As part of the validation of the chemoinformatics module of FunGIMS, the chemical space of the polyamine pathways in Plasmodium was explored in order to obtain a library of molecules based on similarities that could be possible lead-like compounds. This was used in a study of compounds against spermidine synthase from Plasmodium falciparum. By means of similarity and substructure searches the chemoinformatics module was able to produce molecules that are structurally similar. Additional filtering enabled a library of related molecules to be obtained, and used in a docking study. Compounds were found with high docking scores, thereby validating the effectiveness and usefulness of the chemoinformatics module of FunGIMS. Copyright
