Global team aim for faster, more effective tuberculosis diagnosis
TB infects nearly 10 million people each year and kills 1.5 million, ranking it as one of the leading causes of death worldwide. The tuberculosis infection is developing resistance to the antibiotic drugs that we have developed to tackle this disease; almost half a million people each year develop multidrug-resistant TB. There is an urgent need to determine which drugs are required for the individual patient, but determining this using current methods involves growing cultures in a microbiology laboratory – which can take over one month.
A potential faster alternative is to perform whole-genome sequencing (WGS), mapping the whole genetic code of a TB bacterium and looking for genes that allow it to resist particular drugs. Oxford researchers, including the Computational Health Informatics (CHI) Laboratory in the Department of Engineering Science, have teamed up to develop this next-generation technology for predicting antibiotic resistance, using genetic sequencing followed by computational statistics & machine learning.
With a $2.2m (£1.53m) grant from the Bill & Melinda Gates Foundation and a £4m ($5.75m) grant from the Wellcome Trust and MRC Newton Fund, the Oxford team aims to build the world’s large dataset of genetic sequence data for tuberculosis bacteria, and use it to develop the computational techniques required to predict antibiotic resistance. This consortium is led by Prof. Derrick Crook, Director of Public Health England and professor in Oxford’s Nuffield Department of Medicine.
David Clifton, who leads the CHI Lab, said, “This five-year research consortium links leading Oxford clinicians and computational scientists to key sites across the world, whereby we aim to bring to change the way in which tuberculosis is diagnosed. The ultimate aim of this project is to deliver a system to collaborating government agencies across the world that can track the evolving antibiotic resistance of tuberculosis bacteria, and which can continually update itself in light of acquiring new data in real-time.”