Data Mining em aplicações de Desenho Racional de Fármacos

Abstract

Drugs are an asset to the health of the entire population, acting on a biomolecule of our body can increase or decrease their function and thus trigger a therapeutic effect. The major problem of drug design is the time spent in the various stages and therefore the high costs, since it takes some fifteen years for their development, adequate equipment and competent professionals. LogCHEM The tool was developed in order to reduce the time spent in the design of a new drug, allowing the chemical to interact with this tool and design new molecules. The tool uses information on existing molecules and used in actual tests to build models that assist the researcher in building a prototype that has the desired biological activity. This work is an extension of this tool using learning algorithms build predictive models based on pre-sorted molecules that enable the construction of new compounds that besides triggering a certain biological activity, are able to pass a set of ADMET testing. These tests are important for safety, performance and influence drug levels and its activity.