Introduction
Antimicrobial resistance (AMR) is recognised to be a major threat to modern medicine. Better and faster diagnostic tests are key to improved treatment of infections. By accurately identifying whether a bacteria or virus (pathogen) is present, we can identify whether antibiotics should be used. Sequencing the DNA from the bacteria or virus allows for quick identification of the pathogen causing the disease and provides information on whether it can be treated or not and if it is similar to other pathogens that have been seen previously.
The BRC is supporting our activities in developing time- and cost-saving methodologies to isolate bacterial or viral DNA directly from a clinical sample, rather than waiting for it to grow. Once DNA is obtained we use whole genome sequencing methods (WGS) to accurately “read” the genetic code of the pathogen. This code is then compared to a database containing changes, or mutations, which are known to cause resistance to certain drugs. Comparing our DNA of interest to the catalogue determines whether any of those mutations are present and therefore whether certain drugs would kill the bacterial or virus and thus successfully treat the infection. Our group is developing and improving computer programs to look at WGS quickly and accurately and also expanding the databases containing mutations that cause resistance to drugs for certain pathogens. This work will lead to new, quicker and innovative ways to diagnose and treat infection within a clinical setting.
We are also establishing the use of computers to look at health records and sequence data. Combining sequence data with electronic health records can revolutionise the treatment and management of infectious diseases in the National Health Service (NHS) and help England’s health system monitor what infectious diseases are present (national surveillance). We have demonstrated this by investigating the epidemiology and impact of COVID-19 in Oxford University Hospitals NHS Foundation Trust.
Current projects
Faster pathogen diagnostic workflows from samples for viruses and bacteria
The extraction, purification and preparation of pathogen DNA directly from clinical samples avoids the timely and costly need to grow the pathogen in the lab, which can take days or even weeks. We have developed robust and reproducible methods to extract Mycobacteria tuberculosis and Neisseria gonorrhoea from clinical samples, exploring how to diagnose the pathogens causing infection directly from prosthetic joints and establishing how to amplify genes in samples with small amounts of DNA. These methods are being refined and expanded to different pathogens, such as other respiratory viruses.
Once DNA has been extracted, we are using Illumina and Oxford Nanopore Technologies sequencing platforms to read the code. We aim to enhance our sample processing and informatics platform to allow accurate analyses of the WGS and we are expanding our catalogue of mutations causing resistance for certain pathogens.
By identifying the pathogen and predicting resistance in a clinically useful timeframe, we will be able to support earlier clinical decision-making. Our work has resulted in improved processing of Mycobacterium tuberculosis, and the enhanced sample processing and informatics pathways will be implemented and run alongside the standard processing through the National Mycobacterium Reference Service.
COVID-19 Research
The theme is supporting the national response to COVID-19 through several streams:
- We have been involved in the continued validation of swabs, viral transport media and kits for national use in hospital diagnostics, regional testing centres and home self-testing.
- We have co-led the development of a high-throughput test to detect COVID-19 antibodies from blood samples. The test is based on the enzyme-linked immunosorbent assay (ELISA) and detects antibodies to SARS-CoV-2. This test is being used by the Office of National Statistics as part of the national COVID-19 Infection Survey, for which theme co-lead Professor Sarah Walker has been appointed as Chief Investigator. The test has additionally been used to support COVID-19 vaccine development, UK Biobank COVID-19 longitudinal surveillance and to carry out surveillance testing of staff at Oxford University Hospitals NHS Foundation Trust.
- In conjunction with Public Health England, we have evaluated a number of lateral flow antigen tests. Using a four-phase assessment process, tests were identified as suitable for community use.
- We are contributing to the evaluation and validation of a new Oxford Nanopore Technology (ONT) platform that targets and amplifies three SARS-CoV-2 genes (LamPORE). The device can add to COVID-19 screening efforts through its ability to process large numbers of samples and produce results with high sensitivity and specificity quickly. We are assisting with the implementation of the LamPORE platform into Oxford University Hospitals NHS diagnostic laboratory as a pilot to determine whether it is suitable for national implementation.
Combined analysis of clinical record and genomic data
By analysing combined genomic data and electronic health records, we aim to improve the management of infectious diseases within the NHS. This has taken shape through the redesign of the data warehouse of electronic health records with the Clinical Informatics and Big Data research theme.
We are supporting COVID-19 efforts through projects investigating the epidemiology and impact of COVID-19. These include improving COVID-19 surveillance in-long term care, periodic testing of healthcare workers to determine if previous infection provides protection against future infection (SIREN study), and supporting linkage of new COVID-19 test results to UK Biobank for research into the epidemiological and human genetic risk factors for severe infection.
Faster informatics processing and Whole Genome Sequencing analysis
The theme is helping develop effective, accurate, free-to-use software tools for assembling and understanding the genomes of pathogens, to be used for clinical diagnostics (e.g. in hospitals) or public health (e.g. government labs or statistical research), anywhere in the world. To do this, we are adopting and creating a set of computational tools including bioinformatics modules (specific to different pathogens), pipelines (to run the correct tools, in the correct order, for each sample), and a platform to run everything in a robust, easy-to-use, secure way – making genetic insights available to non-experts. At a technical level, our tools handle quality control, genome assembly, species identification, picking up on genetic mutations/variants, prediction of antimicrobial resistance or other clinical features, nearest genetic neighbour detection, downloadable reports, online data dashboards, and the ability to tie these results in with other public data.
Bringing together the results of the Scalable Pathogen Pipeline Platform (SP3), and the Comprehensive Resistance Prediction for Tuberculosis: an International Consortium (CRyPTIC) project, we are now able to go from a tuberculosis sample taken from a patient, to specific recommendations for which drugs may or may not work for that patient, and how that patient’s infection is related to other infections within an outbreak. We have also developed a pipeline for analysing COVID, which can quickly flag Variants of Concern and pick up on newly emerging mutations within minutes: this technology is being rolled out for the world, as the Global Pathogen Analysis System. We will be expanding our tools to cover more pathogens in the coming months and years.
The collaboration includes the Oxford BRC, United Kingdom Public Health Security Agency, Public Health Wales, Public Health Scotland, Animal and Plant Health Agency, the European Bioinformatics Institute, Oracle, and the University of Cardiff, as well as a wide international community of expert users.
Watch
Lecture by Derrick Crook, Professor of Microbiology and Consultant Physician, about how our hospitals are working hard to reduce infection and what that means for patients and healthcare staff. He explains how his team are tracking patients and the history of their treatment over time in the hospital by retrieving and using old records. They are also tracking the progress of different bugs over a number of years.