- Title Information
- Title
- SynerGNN: A Graph Neural Network for Predicting Antibiotic Synergy
- Type of Resource (primo)
- dissertations
- Name:
Personal
- Name Part
- Small, Colin Robert
- Role
- Role Term:
Text
- creator
- Name:
Personal
- Name Part
- Crawford, Lorin
- Role
- Role Term:
Text
- Advisor
- Name:
Personal
- Name Part
- Schell, Jacquelyn
- Role
- Role Term:
Text
- Reader
- Name:
Personal
- Name Part
- Rice, Louis
- Role
- Role Term:
Text
- Reader
- Name:
Corporate
- Name Part
- Brown University. Biology and Medicine: Biotechnology
- Role
- Role Term:
Text
- sponsor
- Origin Information
- Copyright Date
- 2023
- Physical Description
- Extent
- viii, 61 p.
- digitalOrigin
- born digital
- Note:
thesis
- Thesis (Sc. M.)--Brown University, 2023
- Genre (aat)
- theses
- Abstract
- The rise of antimicrobial resistant (AMR) bacteria is one of the world's most pressing public health crises, and our fight against such AMR bacteria is hampered by the time-, labor-, and cost-intensities of discovering novel antibiotics. However, drug repurposing paradigms offer us the opportunity to forgo many of these costs by repositioning existing drugs from their original medicinal purpose into being antibiotic adjuvants, or compounds that when prescribed alongside antibiotics that produce a more potent antibiotic effect than the application of the antibiotic alone. Through the integration of modern deep learning techniques and massive amounts of activity data produced via high-throughput screening of antibiotics and potential adjuvants, we can effectively train models to discovery and predict these novel synergistic antibiotic combinations.
Towards that goal, we propose SynerGNN, a graph neural network model for predicting and discovering novel synergistic antibiotic combinations through screening repurposable molecules. In this work, we evaluate its performance on both antibiotic and non-antibiotic activity datasets and take strides to understand the compounds that it predicted to have the highest antibiotic synergies. We also highlight the challenges of working with limited antibiotic synergy datasets and justify the need for a more complete dataset of antibiotic synergy data the implementation of machine learning techniques that will help us interpret the molecular structures that SynerGNN associates with synergistic activity. Ultimately, models such as SynerGNN will serve as important tools in the continued fight against antimicrobial resistant bacteria in accelerating the screening process of potential antibiotic adjuvants and in discovering novel mechanisms of synergistic action that can be exploited to completely break through existing bacterial resistance mechanisms.
- Subject (fast)
(authorityURI="http://id.worldcat.org/fast", valueURI="http://id.worldcat.org/fast/00817247")
- Topic
- Artificial intelligence
- Subject (fast)
(authorityURI="http://id.worldcat.org/fast", valueURI="http://id.worldcat.org/fast/00810420")
- Topic
- Antibiotics
- Subject (fast)
(authorityURI="http://id.worldcat.org/fast", valueURI="http://id.worldcat.org/fast/02032663")
- Topic
- Deep learning (Machine learning)
- Language
- Language Term (ISO639-2B)
- English
- Record Information
- Record Content Source (marcorg)
- RPB
- Record Creation Date
(encoding="iso8601")
- 20230602