Partnerships

While wastewater-based epidemiology (WBE) has proven capability for public health monitoring in human populations, its application to animal production systems remains largely unexplored. This project addresses this gap by establishing the use of WBE for AMR surveillance in abattoir effluent from poultry, sheep, and cattle processing facilities.

A project is underway to showcase two possible uses of de novo peptide binders in lateral flow immunoassays (LFIA): 1) A new methodology for the rapid development of LFIAs without the use of animal models and 2) The ability to target specific epitopes on antigens for the future development of LFIAs.

The team are aiming to develop a proof-of-concept LFIA to detect beta-lactamase from a range of animal samples. The developed LFIA aims to inform clinical decision making (on use of appropriate antimicrobials) as well as being a valuable research tool for investigating mechanisms of AMR resistance (alongside genomic analysis).

This partnership aims to study the group of babies often forgotten - those who received antibiotics but were not proven to have infection. The project aims to assess the trajectory of the gut microbiome and infant health longitudinally, after exposure to antibiotics in the first days of life – helping understand outcomes and better quantify changes in the microbiome.

Focusing on the Neonatal Intensive Care Unit at UCLH, this project examines the impacts of empirical antibiotic prescribing on infants. By analysing sequencing data and patient health metrics, the study seeks to improve early diagnosis and reduce unnecessary antibiotic use.

This project seeks to develop a global database for source attribution of microbial infections, leveraging machine learning to identify origins of infections rapidly. This initiative will create a comprehensive resource to aid in the fight against AMR across One Health disciplines.

Initial empiric antibiotic choices for the management of bacterial infections are informed by local guidelines that rely on an understanding of local phenotypic AB resistance patterns. The University of Bristol have developed a standalone application that enables straightforward import of local pathology data, and once trained, decision support comparisons between clinician-entered details. In this new collaboration, the team will deploy and validate the app through UCLH’s Experimental Medicine Application Platform (EMAP) - a unique new clinical deployment environment for hosting clinical decision support software tools.  The collaboration will enable rapid route to deployment of the application on NHS systems and provide direct feedback from UCLH clinical microbiologists.

This project aims to harness recent advances in machine learning and deep learning methodologies to establish effective solutions for E. coli source attribution in both animals and humans. By analysing large datasets of E. coli genomes from animal, human, and food isolates, the team will refine existing computational models to probabilistically attribute bacterial isolates to their potential host reservoirs as well as identify promiscuous lineages that pose a more significant risk in terms of anthroponotic or zoonotic dissemination.