The School of Computer Science is pleased to present…
Similarity-Based Graph Neural Network Method for Drug Adverse Side Effect Prediction
MSc Thesis Proposal by:
Mayank Kumar
Date: Monday, August 7,2023
Time: 11:00 am – 12:30 pm
Location: Essex Hall Room 122
Reminders: 1. Two-part attendance mandatory (sign-in sheet, QR Code) 2. Arrive 5-10 minutes prior to event starting - LATECOMERS WILL NOT BE ADMITTED. Note that due to demand, if the room has reached capacity, even if you are "early" admission is not guaranteed. 3. Please be respectful of the presenter by NOT knocking on the door for admittance once the door has been closed whether the presentation has begun or not (If the room is at capacity, overflow is not permitted (ie. sitting on floors) as this is a violation of the Fire Safety code). 4. Be respectful of the decision of the advisor/host of the event if you are not given admittance. The School of Computer Science has numerous events occurring soon
Abstract:
Drug adverse side-effects have substantial impacts on public health, healthcare costs, and drug discovery processes. Hospital admissions and emergency department visits are frequently attributed to adverse drug reactions (ADRs), incurring significant expenses. As medication usage continues to rise, effective management of drug side-effects becomes increasingly crucial. This study aims to predict associations between drugs and adverse side effects using the Drug-Adverse Side-Effect Network (DSN). Traditional experimental methods for determining drug side-effect associations are costly, and time-consuming, necessitating the need for alternative approaches. Our approach frames the problem as a binary classification task within the context of link prediction on a graph using Heterogenous Graph Neural Network. By analyzing the DSN, we seek to identify potential links between drugs and side-effects which in turn aids in early adverse side effect identification, drug development, and patient monitoring. Our preliminary experimentation using Heterogeneous Graph Neural Network, with each drug represented using their PubChem features, on DSN created using SIDER4 shows competitive results and score 0.80 AUC, 0.64 AUPR, and 0.74 F1 over 3- fold setting.
Thesis Committee:
Internal Reader: Dr. Jianguo Lu
External Reader: Dr. Majid Ahmadi
Advisor: Dr. Alioune Ngom
