How to Use Deep Learning to Predict Drug Target Interactions

In silico drug discovery approaches are becoming popular and cost-effective ways of identifying novel therapeutic candidates. In this learning experience, Akshay Bareja, D.Phil. will lead us through a demonstration of how to use deep learning to predict drug-protein interactions. Specifically, we will recreate a published model called "DeepDTA" that only uses drug and protein sequence data (in the form of SMILES strings and amino acid sequences respectively) to predict binding affinity.