Diterepene Molecule Classification

2023
Data Science Project

Tools & Technologies

Python, TensorFlow

Description

Diterpenes are part of the terpene family, which are organic compounds with a carbon skeleton and the formula (C5H8)n. These compounds can be found in certain plant oils and often have important medicinal properties.

NMR spectroscopy is a method for studying a compound’s chemical structure, through a process called structure elucidation. This process exploits the fact that atomic nuclei have certain magnetical moment induced by their spin, making the molecules susceptible to magnetic fields.

This technique measures the absorption of electromagnetic radiation in the spectrum from 4-900 MHz. The resonance frequencies and the number of peaks are used in this machine learning project in conjuction with grid search, knn, and neural networks.