Grant: Master's Award: Canada Graduate Scholarships
University of British Columbia
$27,000.00
chAMP: a machine-learning model for AMP clinical trial outcome prediction
FY 2024-25
Summary
Antimicrobial resistance (AMR) is a global health crisis driven by the misuse of antibiotics. Antimicrobial peptides (AMPs), naturally occurring components of the innate immune system, offer a promising alternative to traditional antibiotics due to their broad-spectrum activity, rapid action against pathogens, and ability to avoid DNA damage in targets. However, AMP toxicity to human cells remains a significant challenge, as seen with the failure of murepavadin's Phase III trial due to kidney toxicity.As of November 2024, ClinicalTrials.gov lists 428 AMP-related trials, of which 348 are inactive or terminated. These studies offer invaluable data on AMP development challenges, failures, and successes.Thus, I propose chAMP;a machine learning model designed to predict AMP clinical trial viability and reasons for failure. By leveraging a comprehensive dataset of the AMPs tested in clinical trials, chAMP will learn sequence embeddings, cluster them by termination reasons, and map new AMPs into the learned space to predict their clinical trial outcomes. This model will be integrated into a pipeline called stAMPede, alongside AMPlify (for de novo AMP discovery) and tAMPer (for toxicity prediction), to streamline AMP development. Outputs will undergo in vitro validation to assess efficacy against priority pathogens and safety towards the host.chAMP reduces reliance on time-intensive and costly experimental workflows, accelerating AMP development and enhancing the arsenal of AMPs against WHO priority pathogens. By equipping researchers with computational tools for AMP prediction, refinement, and validation, this project aims to advance antimicrobial drug development and combat the growing AMR threat.
Department
Program
Master's Award: Canada Graduate Scholarships
Recipient
Location
British Columbia, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database
As published
Id
530156
Name
Zhang Hai Yang E
Region
British Columbia
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
chAMP: a machine-learning model for AMP clinical trial outcome prediction
Programtype2
Studentship Programs
Orgnameinp2
University of British Columbia
Cihramount2
$27,000
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Scholarships
Keyworddelim
Antimicrobial Peptides; Antimicrobial Resistance; Clinical Trials; Data Science; Deep-Learning; Genomics; Immune Response; In Silico Analysis; Software Development; Software Pipeline
Abstract
Antimicrobial resistance (AMR) is a global health crisis driven by the misuse of antibiotics. Antimicrobial peptides (AMPs), naturally occurring components of the innate immune system, offer a promising alternative to traditional antibiotics due to their broad-spectrum activity, rapid action against pathogens, and ability to avoid DNA damage in targets. However, AMP toxicity to human cells remains a significant challenge, as seen with the failure of murepavadin's Phase III trial due to kidney toxicity.As of November 2024, ClinicalTrials.gov lists 428 AMP-related trials, of which 348 are inactive or terminated. These studies offer invaluable data on AMP development challenges, failures, and successes.Thus, I propose chAMP;a machine learning model designed to predict AMP clinical trial viability and reasons for failure. By leveraging a comprehensive dataset of the AMPs tested in clinical trials, chAMP will learn sequence embeddings, cluster them by termination reasons, and map new AMPs into the learned space to predict their clinical trial outcomes. This model will be integrated into a pipeline called stAMPede, alongside AMPlify (for de novo AMP discovery) and tAMPer (for toxicity prediction), to streamline AMP development. Outputs will undergo in vitro validation to assess efficacy against priority pathogens and safety towards the host.chAMP reduces reliance on time-intensive and costly experimental workflows, accelerating AMP development and enhancing the arsenal of AMPs against WHO priority pathogens. By equipping researchers with computational tools for AMP prediction, refinement, and validation, this project aims to advance antimicrobial drug development and combat the growing AMR threat.
Competitiondate
202412
Approvedterm2
1 yr 0 mth
Pinamesdelim
Zhang, Hai Yang E
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $27,000~ $0~