Project: A digital phagogram for personalized phage therapy
2019-11-01 – 2023-10-31
- Abstract
Pathogenic bacteria increasingly become resistant to antibiotics, leading to a decreasing number of therapeutic options, with even no options in case of pandrug-resistant strains. The increasing number of untreatable bacterial infections has sparked a renewed interest in phage therapy in the Western world. Phage therapy is the therapeutic use of phages (viruses that infect bacterial cells) against bacterial infections. Belgium has recently approved a new regulatory framework as the first country worldwide, enabling tailor-made phage products to be used in magistral preparations for treatment of individual patients. Today, the development process of phage-based biologicals remains labor-intensive and costly. In my PhD project, I will focus on developing machine learning tools to predict bacteria-phage interactions at the strain level and creating digital phagograms based on experimental validation (equivalent to an antibiogram for antibiotics). Machine learning algorithms will learn from known interactions between bacteria and phage proteins to predict new interactions. This enables rapid selection and production of appropriate phages, which significantly increases the speed at which phages are characterized. Our research units will work with the Queen Astrid military hospital (see letter of support) to incorporate this framework in their phage therapy approach. In this way, my PhD project further strengthens Flanders’ pioneering role in this antibacterial strategy.
-
- Journal Article
- A1
- open access
Hyperdimensional computing : a fast, robust, and interpretable paradigm for biological data
-
- Journal Article
- A1
- open access
DepoScope : accurate phage depolymerase annotation and domain delineation using large language models
-
- Journal Article
- A1
- open access
Prediction of Klebsiella phage-host specificity at the strain level
-
- Journal Article
- A1
- open access
CkP1 bacteriophage, a S16-like myovirus that recognizes Citrobacter koseri lipopolysaccharide through its long tail fibers
-
- Conference Paper
- C3
- open access
Two novel approaches to identify phage receptor-binding protein sequences
-
Machine learning to assist in large-scale, activity-based synthetic cannabinoid receptor agonist screening of serum samples
-
- Journal Article
- A1
- open access
Identification of phage receptor-binding protein sequences with hidden Markov models and an extreme gradient boosting classifier
-
- Journal Article
- A1
- open access
The specific capsule depolymerase of phage PMK34 sensitizes Acinetobacter baumannii to serum killing
-
Digital phagograms : predicting phage infectivity through a multilayer machine learning approach