Using deep learning in protein interactions to detect emerging disease threats  

A digital illustration of an abstract form representing a network

There is a diagnostics challenge in detecting foodborne and disease-causing strains of Shigatoxigenic Escherichia coli. There are multiple serotypes and they are genetically diverse which means it can be tricky to identify those that cause disease.  

A project led by Scotland’s Rural College, working alongside the Science and Technology Facilities Council and PrimerDesign Ltd is using bioinformatics to help classify these strains. The project hopes to inform potential application towards the next generation of diagnostic tools that use sequence-based differences to identify pathogens.  Overall, the project funded by the Food Safety Research Network aims to help detect bacteria that are potentially pathogenic and classed as high risk to human health.