MSc Thesis Proposal Announcement of Yakin Patel: "Machine Learning Approach to Drug Repurposing with SELFIES"

 

Date: Tuesday October 11th, 2022 

Time:  10:00am – 11:30am 

Location: Essex Hall, Room 122 

Reminder: Two-part attendance required: Part I (Scan the QR code, fill in online form) and Part II (sign in sheet). *Make sure you are in the room at least 5-10 minutes BEFORE the presentation starts – once the presentation has begun, latecomers will not be admitted.

 

Drug repurposing is a promising and efficient way for discovering new uses from existing drugs. Identifying drug-protein interactions has been a key step in drug discovery and drug repurposing. Many computational methods have been proposed to predict whether the drug and the protein will interact or not (Binary Classification). Drug-Protein binding affinity is another type of data which could show strength of the bond between the drug and protein. DeepDTA, Co-VAE and few other models have been presented for predicting drug protein binding affinity and generate new pairs. However, all the models have used SMILES, drug representation that has few flaws when used with generative model. Here we propose a model that uses VAE to generate new drug-protein pair and use SELFIES drug representation that is found to be 100% robust when compared with SMILES and produces more valid molecules. 

 

Keywords: Drug-Protein Interaction, Variational Autoencoder, SMILES, SELFIES 

Internal Reader: Dr. Sherif Saad           

External Reader: Dr. Majid Ahmadi 

Advisor: Dr. Alioune Ngom