INDCAL-LOCAL

DESCRIPTION

The motive of the project is to calculate various types of indices in the development of quantitative structure-activity relationships in which the biochemical activity or other properties of molecules are correlated with their chemical structure. This is being applied in many disciplines for example risk assessment, toxicity prediction, and regulatory decisions in addition to drug discovery and lead optimization. Man-made research tends to mishappening and accuracy of the results aren’t as expected. But when machines are trained to calculate the indices of molecules, the properties of the unknown chemical structure can be easily identified. The input would be the SMILES notation of chemical compounds. SMILES is SIMPLIFIED MOLECULAR INPUT LINE SYSTEM, which is used to translate a chemical’s threedimensional structure into a string of symbols that is easily understood by computer software. Once it’s provided, topological indices like generalized randic index, augmented zagreb index, geometric arithmetic index, harmonic index, product connectivity index, general sum-connectivity index, and atom-bond connectivity index are employed to calculate the bioactivity of chemicals. Once the indices are found, it would be the training set of the machine learning model that is to be trained. Once trained, the model can predict the properties of unknown chemical compounds on it’s own. This would be the future enhancement of the project