Computational approaches in drug chemistry leveraging python powered QSPR study of antimalaria compounds by using artificial neural networks.

The application of Machine Learning has become a revolutionary instrument in the domain of pharmaceutical research. Machine learning enables the modelling of Quantitative Structure Property Relationship, a crucial task in forecasting the physiochemical characteristics of drugs. In this study we utilized machine learning algorithms namely Artificial Neural Networks and Random Forest to predict physiochemical characteristics of Anti-malaria drugs. These models utilize several topological indices global variables quantifying the connectivity and geometric characteristics of molecules to estimate the ability of prospective antimalarial compounds to interact with the target enzyme and other physicochemical parameters. Molecular descriptors such as size, shape, and electronic structure indices are a way of mapping molecular properties into a set of quantitative data that can be analyzed by Machine Learning techniques. By carrying out regression analysis with the help of Artificial Neural Networks and Random Forest, the corresponding changes in the molecular structures and their effects on effectiveness and properties of the potential drugs can be predicted, thereby supporting the search for new therapeutic compounds. Machine learning not only observe the drug development process but also facilitates to look at chemical datasets with respect to high order non-linear relationship, which are essential to improve antimalarial drug candidates and pharmacokinetic properties.