Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Master's Award: Canada Graduate Scholarships

McMaster University

$27,000.00

Discovering synergistic combination therapies against Acinetobacter baumannii using machine learning approaches

FY 2024-25

Summary
Antimicrobial resistance (AMR) is expected to cause 10 million deaths per year by 2050. Among the largest threats is carbapenem-resistant Acinetobacter baumannii (CRAB), a gram-negative nosocomial pathogen on the WHO's Priority Pathogens list. New treatments against CRAB are urgently needed. Drug repurposing - the application of approved drugs to novel clinical indications - presents a new strategy for discovering clinically relevant antibiotics since these drugs already have validated safety profile. Using these compounds in combination holds great promise for developing treatments for CRAB. Combining them can slow the development of resistance and allow lower doses to be administered, reducing possible side effects. However, screening for synergistic combinations is challenging, as current in vitro methods cannot explore vast combinatorial spaces. Alternatively, ML methods can efficiently explore billions of combinations in silico in just days. Indeed, this project aims to leverage ML approaches to predict novel synergistic chemical-chemical combinations of repurposed clinical drugs against CRAB. I began by processing a dataset of chemical-chemical combinations tested for growth inhibition against CRAB. Then, I trained a ML model to predict novel synergies. As proof of concept, I ran predictions on 1 million random combinations from an in-house drug library of 11,000 molecules (out of 55M possible) and am currently validating the top 50 combinations in the lab. Moving forward, I will extend predictions to all 55M possible combinations, selecting 200 novel combinations for further testing. The most potent combinations will be further explored to assess their clinical potential.
Program
Master's Award: Canada Graduate Scholarships
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database

As published

Id
529641
Name
Ta Wesley
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
Discovering synergistic combination therapies against Acinetobacter baumannii using machine learning approaches
Programtype2
Studentship Programs
Orgnameinp2
McMaster University
Cihramount2
$27,000
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Scholarships
Keyworddelim
Acinetobacter Baumannii; Antibiotic; Antibiotic Resistance; Artificial Intelligence; Chemistry; Combination Therapies; Drug Discovery; Machine Learning
Abstract
Antimicrobial resistance (AMR) is expected to cause 10 million deaths per year by 2050. Among the largest threats is carbapenem-resistant Acinetobacter baumannii (CRAB), a gram-negative nosocomial pathogen on the WHO's Priority Pathogens list. New treatments against CRAB are urgently needed. Drug repurposing - the application of approved drugs to novel clinical indications - presents a new strategy for discovering clinically relevant antibiotics since these drugs already have validated safety profile. Using these compounds in combination holds great promise for developing treatments for CRAB. Combining them can slow the development of resistance and allow lower doses to be administered, reducing possible side effects. However, screening for synergistic combinations is challenging, as current in vitro methods cannot explore vast combinatorial spaces. Alternatively, ML methods can efficiently explore billions of combinations in silico in just days. Indeed, this project aims to leverage ML approaches to predict novel synergistic chemical-chemical combinations of repurposed clinical drugs against CRAB. I began by processing a dataset of chemical-chemical combinations tested for growth inhibition against CRAB. Then, I trained a ML model to predict novel synergies. As proof of concept, I ran predictions on 1 million random combinations from an in-house drug library of 11,000 molecules (out of 55M possible) and am currently validating the top 50 combinations in the lab. Moving forward, I will extend predictions to all 55M possible combinations, selecting 200 novel combinations for further testing. The most potent combinations will be further explored to assess their clinical potential.
Competitiondate
202412
Approvedterm2
1 yr 0 mth
Pinamesdelim
Ta, Wesley
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $27,000~ $0~