Grant: Master's Award: Canada Graduate Scholarships
McMaster University
$17,500.00
Generative AI-guided design of antibiotics that inhibit siderophore uptake in Pseudomonas aeruginosa
EF 2023-24
Résumé
There is an urgent need for novel antibiotics targeting Pseudomonas aeruginosa, a deadly Gram-negative pathogen responsible for a significant number of deaths due to antibiotic resistance. The study aims to leverage generative AI, specifically, Full-Atom Iterative Refinement (FAIR), to design proteins that effectively bind to the siderophore pyoverdine, a crucial iron chelator vital to P. aeruginosa pathogenesis. The approach involves utilizing diffusion models, a form of generative AI, to navigate the vast protein space and iteratively denoise known protein-ligand complexes, thereby training a neural network (NN) to learn key protein features essential for binding. FAIR, specifically designed for ligand-specific protein pockets, addresses limitations in other generative models, offering improved side-chain modelling and ligand flexibility. The trained NN acquires essential features for ligand binding, including residue and atom interactions. The objective is to generate proteins that bind to pyoverdine, preventing iron uptake and mitigating virulence with a lower frequency of resistance. The study plans to express and purify the generated proteins, determining their binding affinity through isothermal titration calorimetry. The most promising proteins will undergo in vitro growth inhibition assays against planktonic P. aeruginosa in an iron-deficient medium, paving the way for subsequent in vivo animal testing. This innovative approach, combining AI-guided protein design with rigorous experimental validation, holds promise for developing effective antibiotics targeting P. aeruginosa and addresses the challenges posed by antibiotic resistance in a comprehensive manner.
Programme
Master's Award: Canada Graduate Scholarships
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
CIHR Funding Decisions Database
Tel que publié
Id
506449
Name
McLellan Stewart J
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
Generative AI-guided design of antibiotics that inhibit siderophore uptake in Pseudomonas aeruginosa
Programtype2
Studentship Programs
Orgnameinp2
McMaster University
Cihramount2
$17,500
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Scholarships
Keyworddelim
Antimicrobial Resistance; Artificial Intelligence; Diffusion Model; Frequency Of Resistance; Iron Chelation; Machine Learning; Neural Network; Protein Design; Pseudomonas Aeruginosa; Siderophore
Abstract
There is an urgent need for novel antibiotics targeting Pseudomonas aeruginosa, a deadly Gram-negative pathogen responsible for a significant number of deaths due to antibiotic resistance. The study aims to leverage generative AI, specifically, Full-Atom Iterative Refinement (FAIR), to design proteins that effectively bind to the siderophore pyoverdine, a crucial iron chelator vital to P. aeruginosa pathogenesis. The approach involves utilizing diffusion models, a form of generative AI, to navigate the vast protein space and iteratively denoise known protein-ligand complexes, thereby training a neural network (NN) to learn key protein features essential for binding. FAIR, specifically designed for ligand-specific protein pockets, addresses limitations in other generative models, offering improved side-chain modelling and ligand flexibility. The trained NN acquires essential features for ligand binding, including residue and atom interactions. The objective is to generate proteins that bind to pyoverdine, preventing iron uptake and mitigating virulence with a lower frequency of resistance. The study plans to express and purify the generated proteins, determining their binding affinity through isothermal titration calorimetry. The most promising proteins will undergo in vitro growth inhibition assays against planktonic P. aeruginosa in an iron-deficient medium, paving the way for subsequent in vivo animal testing. This innovative approach, combining AI-guided protein design with rigorous experimental validation, holds promise for developing effective antibiotics targeting P. aeruginosa and addresses the challenges posed by antibiotic resistance in a comprehensive manner.
Competitiondate
202312
Approvedterm2
1 yr 0 mth
Pinamesdelim
McLellan, Stewart J
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $17,500~ $0~