In silico metodologija u toksikologiji - softveri za predviđanje toksičnosti

Abstract

For over a century, one of the main toxicologists' goals was finding the relationship between the chemical structure and toxicity. The development of computer technologies has lead to the production of software models that could be applied for this purpose. However, use of in silico methodology in toxicology is very complex. Having in mind that toxic effect occurs through multiple mechanisms on different tissues and organs, creating software that can make reliable toxicity predictions is very complicated. In silico methods found the application in toxicity assessment of substances for which data obtained from traditional toxicity tests is lacking (metabolites, drug impurities, etc.). Also, in silico methods can be used in the process of chemical synthesis, with the aim of designing safer chemicals. Currently, in silico methods for toxicity predictions are most commonly used along with standard toxicological tests, with a tendency to replace in vitro and in vivo toxicity tests in the future. This paper will present the strengths and limitations of commonly used software in in silico toxicology (OECD QSAR Toolbox, Derek Nexus, ToxTree i MultiCASE).