Graph Neural Networks for drug design – UROP Spring Symposium 2024

Graph Neural Networks for drug design

Isabella Chittaro

Pronouns: she/her

Research Mentor(s): Cristian Minoccheri
Research Mentor School/College/Department: Computational Medicine and Bioinformatics / Medicine
Program:
Authors: Cristian Minoccheri, Isabella Chittaro, Mihir Arya, Rinny Fan, George Simmons
Session: Session 6: 3:40 pm – 4:30 pm
Poster: 30

Abstract

The diminishing efficacy of antibiotics against increasingly resistant bacteria poses a significant public health challenge. Despite the escalating threat of antibiotic resistance, the development of new antibiotics is hindered by scientific, regulatory, and financial barriers. Recent literature suggests that antimicrobial polymers, despite their stochastic nature, may be viable alternatives to traditional peptides due to lower cost and production efficiency. To facilitate this process, we introduce attention to a directed, weighted Message Passing Neural Network (MPNN) developed for polymer representations as directed graphs. We further investigate the use of classifiers trained as energy models to generate novel data for research and development of antimicrobial polymers.

Biomedical Sciences, Interdisciplinary, Natural/Life Sciences

lsa logoum logo