An application of some machine learning methods for biological data modeling

The development of fast and reliable methods for predicting the biological activity of the substances in computational biology is of a great importance. This improves the development of some new compounds while keeping costs low. Among many scientists, an attractive target for docking experiments is the Delta-opioid receptor (DOR) and delta-opioid ligands (DOL). Their biological efficacy can be measured by various techniques, which could facilitate the establishment of the relationship between the structure of the compounds and their biological effect. The relationship between the results of the computer experiments and the biological activity of these compounds is modelled by using machine learning regressors. The primary goal of this study is to determine the most appropriate neural network for modelling the relationship between in vitro and in silico results for DOR and delta-opioid ligands.