Project Description
Peptide-based therapeutics are an emerging class of drugs that represent roughly 8% of the FDA approved drugs in the last 10 years. The objective of this project is to build a machine learning model that can accurately predict the binding free energy (ΔG) for two interacting proteins to aid with the drug discovery process. This will be achieved through analysis of numerous publicly available protein datasets, data provided from molecular dynamic simulations, and physical/structural data calculated using Rosetta. The results will be validated by comparing the mean absolute error of our results to both experimentally validated binding energies as well as other published machine learning model predictions.
Tasks and Responsibilites
The student will be primarily responsible for benchmarking other published machine learning models to provide data to compare our model to. Their other responsibilities include using Rosetta to model and extract physical data from protein structures as well as writing code to help consolidate and organize the massive amount of data that will eventually be fed into our machine learning model for training purposes.
Desired Qualifications
None Listed.