Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: DRA : Canada Graduate Research Scholarship - Doctoral Program (CGRS D)

McMaster University

$120,000.00

The ESKAPE model: open-source antibiotic prediction for everyone

EF 2025-26

Résumé
The global dissemination of antibiotic resistance is reducing available treatment options for bacterial infections. The breadth of this crisis is demonstrated by the ESKAPE pathogens - a group of bacteria that are responsible for the majority of hospital-acquired infections and resistant to last resort antibiotics. Alarmingly, a new class of antibiotics has not emerged for clinical use since 1987 due to bottlenecks in our current discovery pipelines and the stark failure rate of antibiotic candidates throughout clinical trials. My work leverages machine learning (ML) to virtually screen vast chemical libraries for antibacterial, non-toxic molecules to facilitate new antibiotic discovery. Virtual screening offers a more efficient way to identify antibiotic leads compared to in vitro screening methods. However, ML often requires a steep learning curve for biological experts. I am working to develop an online ML-based antibiotic discovery platform called the ESKAPE model that is accessible to everyone regardless of computational skill level. This resource contains ML models that predict the antibacterial activity of molecules against each ESKAPE pathogen, and toxicity against five human cell lines that span the kidneys, liver, lungs, heart and skin. To date, I have trained the antibacterial prediction models and used them to discover three antibacterial molecules that have similar potencies to current clinical antibiotics. Moving forward, I will develop the toxicity models and use them alongside the antibacterial models to identify molecules that are antibacterial and non-toxic to demonstrate how my models can be leveraged in combination for antibiotic discovery. The ESKAPE model is currently available at eskape.mcmaster.ca with all antibacterial models. By providing free access to this tool, I aim to lower the technical barrier to ML adoption and broaden the capacity for the antibiotic discovery community to discover antibiotics more efficiently and at greater scale.
Programme
DRA : Canada Graduate Research Scholarship - Doctoral Program (CGRS D)
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
CIHR Funding Decisions Database

Tel que publié

Id
559295
Name
Arnold Patricia A
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Infection and Immunity
Theme2
Biomedical
Prcname2
DT1 (Biomedical and/or Clinical Research) CGRS Doctoral
Projecttitle
The ESKAPE model: open-source antibiotic prediction for everyone
Programtype2
Studentship Programs
Orgnameinp2
McMaster University
Cihramount2
$120,000
Cihrequipment2
$0
Programname2
DRA : Canada Graduate Research Scholarship - Doctoral Program (CGRS D)
Keyworddelim
Antibiotic Resistance; Antibiotics; Artificial Intelligence; Biochemistry; Drug Discovery; Machine Learning; Microbiology; Virtual Screening
Abstract
The global dissemination of antibiotic resistance is reducing available treatment options for bacterial infections. The breadth of this crisis is demonstrated by the ESKAPE pathogens - a group of bacteria that are responsible for the majority of hospital-acquired infections and resistant to last resort antibiotics. Alarmingly, a new class of antibiotics has not emerged for clinical use since 1987 due to bottlenecks in our current discovery pipelines and the stark failure rate of antibiotic candidates throughout clinical trials. My work leverages machine learning (ML) to virtually screen vast chemical libraries for antibacterial, non-toxic molecules to facilitate new antibiotic discovery. Virtual screening offers a more efficient way to identify antibiotic leads compared to in vitro screening methods. However, ML often requires a steep learning curve for biological experts. I am working to develop an online ML-based antibiotic discovery platform called the ESKAPE model that is accessible to everyone regardless of computational skill level. This resource contains ML models that predict the antibacterial activity of molecules against each ESKAPE pathogen, and toxicity against five human cell lines that span the kidneys, liver, lungs, heart and skin. To date, I have trained the antibacterial prediction models and used them to discover three antibacterial molecules that have similar potencies to current clinical antibiotics. Moving forward, I will develop the toxicity models and use them alongside the antibacterial models to identify molecules that are antibacterial and non-toxic to demonstrate how my models can be leveraged in combination for antibiotic discovery. The ESKAPE model is currently available at eskape.mcmaster.ca with all antibacterial models. By providing free access to this tool, I aim to lower the technical barrier to ML adoption and broaden the capacity for the antibiotic discovery community to discover antibiotics more efficiently and at greater scale.
Competitiondate
202511
Approvedterm2
3 yrs 0 mth
Pinamesdelim
Arnold, Patricia A
Approvedterm
3 ans 0 mois
Cihrcontribution2
~ $120,000~ $0~