COG-UK research collaborations

From the outset of COG-UK, it was clear that its contribution to the pandemic lay not only in its ability to scale up sequencing of the SARS-CoV-2 virus genome, but also in the analysis researchers could carry out on the data. Some of this could be done immediately, but it also required investment in research to realise the full potential of the data. In order to do this, it set up its own internal Research Project Fund to support investigations that could help understand the biology of the SARS-CoV-2 virus, its behaviour and spread as well as the impact of various public health measures.

The first round of research funding was awarded in June 2021 to 12 consortium partners, including 11 universities across the UK and one Public Health Agency. A year later seed funding was awarded to seven early career researchers who 'had worked in COG-UK and whose own research plans had been affected during the pandemic.' Four new research projects around metagenomics were additionally given funding (COG-UK Research). Table 1 lists the key research projects funded by COG-UK. As well as COG-UK funding and conducting its own research projects, it also participated in a number of national studies (table 2).

More details are given below for three projects that COG-UK was involved in: Hospital-Onset COVID-19 Infections (HOCI study), Genetics of Mortality In Critical Care (GenOMICC study) and wastewater sampling.

Table 1: Core research studies sponsored by COG-UK

Type	Study
SARS-CoV-2 variants	MRC-University of Glasgow Centre for Virus Research - observational study to monitor vaccine response in the Scottish population following exposure to SARS-CoV-2 variants.
SARS-CoV-2 variants	Northumbria University - develop new computational methods for Variant of Concern detection and Variant of Interest classification.
SARS-CoV-2 variants	Queen’s University Belfast- study SARS-CoV-2 genomic variation within individuals to help understand vaccine evasion and any evidence of intra-host evolution will be examined.
SARS-CoV-2 variants	University of Sheffield - examine the impact of SARS-CoV-2 mutations on recognition by a specific type of immune cell (T-cells).
SARS-CoV-2 variants	University College London - use a genomic approach to identify potential reservoirs for the emergence of SARS-CoV-2 variants of concern.
SARS-CoV-2 variants	University of Nottingham - develop and deploy a sequencing pipeline for seasonal human coronaviruses to monitor for recombination with SARS-CoV-2 and post-pandemic evolution.
Targeting treatments	University of Oxford - develop means to move from pandemic response to pandemic preparedness through the integration of diagnostics, rapid genotyping and whole genome sequencing of SARS-CoV-2 and other respiratory pathogens.
Genome quality control	Quadram Institute - develop an online resource for SARS-CoV-2 genome quality control, improving data quality over and above standard QC processes. The tool will be designed so it can be repurposed for other pathogens.
Integration of genomes and metadata	Public Health Agency Northern Ireland - improve the integration of genomes and contract tracing data for outbreak control.
Integration of genomes and metadata	University of Liverpool - link data from COG-UK and Cheshire and Merseyside’s Combined Intelligence for Public Health Action dashboard.
Integration of genomes and metadata	University of Exeter - implement models to infer transmission directionality during hospital-associated infection.
Integration of genomes and metadata	University of Portsmouth - integrate viral genomics data from COG-UK and from the Sequencing and Tracking of Phylogeny in COVID-19 programme (STOP COVID-19) to advance understanding of COVID-19 clinical severity and nosocomial transmission on the south coast.
Metagenomics	MRC-University of Glasgow Centre for Virus Research - study to identify viruses causing undiagnosed febrile illness in returning travellers to the UK using enhanced NGS-based metagenomic methods .In partnership with the Rare and Imported Pathogens Laboratory (from UK Health Security Agency), using undiagnosed samples from returning travellers to the UK, the research team will evaluate the capacity for emerging metagenomic methods to enhance the diagnosis of viruses causing disease in this important risk group.
Metagenomics	University of Sheffield - establish a new method to enhance efficient recovery of viral genomes using Oxford Nanopore Technology sequencing devices, and characterise the virome in immunosuppressed and critically ill patients.
Metagenomics	University of Oxford - test and optimise approaches for virus metagenomics on portable sequencing platforms, improving enrichment of viral reads using on-device selective sequencing. Aim to advance approaches for faster and more sensitive detection of emerging viral outbreaks globally.
Metagenomics	University of Nottingham - develop diagnostic metagenomics for helping with patients hospitalised with symptomatic undiagnosed respiratory illness, including critically ill cohorts requiring intensive high dependency care.
Early career funding scheme	Early career scientists could apply for up to £25,000 funding for individual research projects. Seven were awarded.

Table 2: National studies in which COG-UK contributed

Project	Scope	Sponsor
COG-UK HOCI (Hospital-Onset COVID-19 Infections) study	This investigated how integrating rapid, real-time COVID-19 genomic sequencing can impact decision-making by infection control teams to prevent the spread of the SARS-CoV-2 virus in NHS hospitals.	COG-UK, UCL
COVID-19 in Prisons (CiPS) Study	Testing staff and inmates at 28 prisons across England.
GenOMICC (Genetics of Mortality In Critical Care) study	Sequencing the genomes of thousands of patients who were severely ill with coronavirus and matching this with virus genome data provided by COG-UK to understand how a person’s genomics may influence their susceptibility to COVID-19.	In partnership with Genomics England, Illumina and the NHS.
Real-time Assessment of Community Transmission (REACT) study	Population surveillance study undertaken in England to examine the prevalence of the virus causing COVID-19 in the general population, involving over 150,000 participants each month.	Commissioned by the Department of Health and Social Care and is being carried out by Imperial College London in partnership with Ipsos MORI and Imperial College Healthcare NHS Trust.
SARS-CoV-2 Immunity & Reinfection Evaluation (SIREN) study	Impact of detectable anti SARS-CoV-2 antibody on the incidence of COVID-19 in healthcare workers to understand whether prior infection with SARS-CoV-2 can protect against future infection with the same virus.	UKHSA, Public Health Scotland, Public Health Wales, Northern Ireland Public Health Agency
Vaccine trial	Test safety and efficacy of ChAdOx1 nCoV-19 vaccine (Oxford vaccine)
Vaccine trial	Test safety and efficacy of NVX-CoV2373 vaccine (Novavax vaccine)
VIVALDI Study (started June 2020, ends 31 March 2023)	Determine how many care-home staff and residents were infected with COVID-19 and inform decisions around the best approach to COVID-19 testing in the future.	Led by UCL in collaboration with Four Seasons healthcare, a large care home chain, and the Department of Health and Social Care.
Wastewater genomic surveillance	Study to evaluate whether genomic sequencing of wastewater samples can provide an alternative means of detecting and tracking virus variants in the population.	Funded by Department of Health and Social Care, UK, Natural Environmental Research Council, UK, COG-UK

Each dataset was augmented by the sequencing data generated through COG-UK.

COG-UK HOCI study

One of the earliest research projects funded by COG-UK was the Hospital-Onset COVID-19 Infections (HOCI) Study. The project was initiated in April 2020 by Professor Judith Breuer at University College London. Her aim was to assess the utility of integrating a report incorporating results from rapid-real-time genome sequencing within the hospital setting for the prevention and control of COVID-19. Previously retrospective studies had shown genome sequencing of epidemic viruses to be more effective at identifying transmission routes than infection prevention control investigations alone. In the case of the HOCI study, Breuer wanted to be much more proactive. According to James Blackstone who helped manage the project, Breuer wanted to 'look at the virus sequences of people while they were still in hospital and while they were still potentially transmitting to others, as vectors of transmission (whether it be staff, whether it be certain places) so that situations could be changed in order to prevent further transmission occurring' (Blackstone transcript).

The HOCI study protocol specifies it was designed to take advantage of 'the COG-UK sequencing platform, with its mixed model of smaller sequencing hubs located close to hospitals and a large centralised hub sequencing most viruses' . It aimed to recruit 2,380 patients admitted to 14 NHS hospitals across the UK over winter-spring 2020 to 2021. To be eligible, patients had to test positive with a SARS-CoV-2 PCR test within 48 hours of admission to hospital, having not been diagnosed or suspected of having the disease when admitted. The project had two key phases: rapid and slow. The rapid one required a sequence report being turned around within 48 hours based on rapid local lab processing where the hospital would be able to conduct the sequencing internally. By contrast, the slow phase aimed to provide a report within 5-10 days as would happen if the sequencing was done outside of a hospital site by a central laboratory. Prior to implementing the trial, each site was expected to have collected eligible patient data for four weeks to characterise its standard infection control practice. This served as a baseline for assessing the impact on infection control of the rapid and slow turnaround of the sequencing report (Blackstone 2021; Blackstone 2022; Stirrup).

Getting the HOCI study up and running encountered several challenges, which were overcome by the summer of 2020. The project was in part delayed because it took time to get contracts in place. Another was the amount of work and time commitment it required from already hard-pressed NHS staff. Breuer points out that 'running the trial in the pandemic was fraught with difficulties as people were working to maximum capacity and every infection control procedure was already being implemented as far as possible' (Breuer transcript). Overall, Blackstone explains the pathway had between 15 and 20 steps. He says that it is not an exaggeration 'to say we had to move heaven and earth to get the rapid phase implemented at the sites in anything like the time we expected.' Even with the best efforts he recalls 'the median rapid turnaround time was about 4 to 5 days, it wasn't the 48 hours we were hoping for. And the standard turnaround time which we were aiming for, 5-10 days, we got about more like 15 days out of it. So everything was harder, everything took longer than was expected' (Blackstone transcript).

According to Blackstone, one of the major bottlenecks to the quick turnaround was the difficulties encountered with collecting patient metadata. Part of the problem was that such data had to be collected manually because most sites were not set up to export it electronically. Just how big a task this was is highlighted by the fact that Blackstone says 'We had 2,000-2,300 index cases and a further 20,000-30,000 additional cases that were potentially linked to those cases.' That meant in a 4-month period the HOCI research teams were collecting supportive data for up to 30,000 patients. The bulk of the workload fell on the shoulders of research nurses or administrators for whom it was a major learning curve (Blackstone transcript).

Looking back, Blackstone recalls one of the pivotal moments for the study happened in November 2020 when the first sequence report came back run on real data. At that point, he states 'I was extremely excited to see the first report come back and see there were potential links here between these patients and these patients, and the chart at the bottom that was illustrating all this. I thought despite all the challenges we'd encountered over the course of the previous seven or eight months, or whatever it was, that this is actually viable' (Blackstone transcript).

GenOMICC study

Very early on in getting the ball rolling with COG-UK, Professor Sharon Peacock was contacted by Professor Sir Mark Caulfield, then the Chief Scientist at Genomics England. Invited by Peacock to COG-UK's first inaugural meeting, Caulfield was interested in setting up a study to investigate how a person's genetics affected their susceptibility to developing severe COVID-19 requiring critical care or hospitalisation. For Caulfield, COG-UK provided an ideal opportunity for doing this work because, as he says, 'it mobilised, in a sort of Dunkirk little ships type way, the entire sequencing infrastructure for pathogen sequencing in both academic and NHS environments.' Essentially, Caulfield says, the idea from the outset was to link the human data with the virus data. One of his inspirations for considering combining the viral and host sequences came from the work conducted with the Ebola virus using genotyping. The advantage now was that they could investigate the viral-host interaction with whole genome sequencing (Peacock transcript; Caulfield transcript).

A separate group was formed within COG-UK to work out how the sequencing data collected about the virus could be merged with human genetic data collected through Genomics England. Set up in 2014 to sequence 100,000 genomes from NHS patients, Genomics England was well placed to carry out a study into COVID-19. Originally founded to sequence patients with rare diseases and cancer, in 2015 Genomics England helped launch an open global consortium of intensive care clinicians to investigate how genetic factors influenced outcomes in intensive care from diseases such as SARS, flu and sepsis. Known as Genetics of Susceptibility and Mortality in Critical Care, or GenOMICC, this group is led by the University of Edinburgh in partnership with Genomics England, Queen Mary, NHS Lothian and the Intensive Care National Audit and Research Centre (Queen Mary London).

By early May 2020, Caulfield and his colleagues had managed to get funding in place from the Department of Health and Social Care, LifeArc and the UK Medical Research Council as well as some assistance from Illumina who provided genomes for free. The team, led by Professor Kenneth Baillie at the University of Edinburgh, set out to identify specific human genes that predispose people to become seriously ill with COVID-19 so as to pave the way to more targeted therapy (BBC). This was a huge undertaking logistically because it involved enrolling nearly 30,000 people in 18 months and what Caulfield calls 'a lot of moving parts within the study to try and make it work'. The project was helped by the fact that Genomics England already had a team in place ready to do such work (Caulfield transcript). So far most of the analysis has been done on data relating to human genetics and work is ongoing to merge the viral genomes with human genomes.

Within the context of the human study this looked at the genetics of more than 57,000 people. This included genomes sequenced from 7,491 critically ill patients from 224 intensive care units in the UK, and 1,630 genomes from people with mild COVID-19. This data was then compared with the DNA of 48,400 participants in Genome England's 100,000 Genomes Project who did not have COVID-19 (Queen Mary London). The first results from the human genetics arm of the project were published in March 2020 (Kousathanas). According to Caulfield the really important thing to come out of the human genetics study was they were able to 'identify potential areas of the genome that could be therapeutic targets' (Caulfield transcript).

Wastewater study

Most of COG-UK's efforts were focused on sequencing samples collected directly from people with COVID-19 within hospitals or in the community. While useful for giving detailed information about the SARS-CoV-2 virus and its variants in individuals, it provided a partial view of the spread of the virus at the community level. One of the ways to get a more comprehensive view was to look at wastewater samples, which can contain SARS-CoV-2 shed in the faeces of people with asymptomatic and symptomatic infection. Already used for many years to detect the prevalence of the polio virus and more recently to study antimicrobial resistance and enteric viruses such as norovirus, hepatitis A and E, adenovirus and rotavirus, the same approach was quickly explored for COVID-19 (COG-UK Jan 2021).

In December 2020, COG-UK held a seminar to explore how sequencing could be used to help identify and monitor new SARS-CoV-2 variants and provide early warnings of an outbreak. Thereafter, a number of COG-UK sites helped sequence wastewater samples. A key scientist in this work was Professor Steve Paterson at the University of Liverpool. Between November 2021 and February 2022, he and his team analysed more than 19,000 samples from 524 wastewater sites across England. Collected twice a week, these samples covered 70 per cent of the sewage from the English population. Through this work Paterson's team demonstrated that wastewater data was able to track the spread of the Omicron variant across England in fine detail and closely aligned with data collected from individual testing. Based on this they concluded that wastewater genomic surveillance could 'reliably fill the monitoring gap left by reduced individual testing in a more affordable way' (Brunner).

References

BBC (13 May 2020) 'Coronavirus: Thousands of Covid patients to have genome studied'.Back

Blackstone, J, Stirrup, O, Mapp, OF, et al (29 March 2022) 'Protocol for the COG-UK hospital-onset COVID-19 infection (HOCI) multicentre interventional clinical study: evaluating the efficacy of rapid genome sequencing of SARS-CoV-2 in limiting the spread of COVID-19 in UK NHS hospitals', British Medical Journal.,12/4.Back

Blackstone, J (14 Oct 2021) 'Challenges of implementing an infection control clinical trial during COVID-19', presentation to COG-UK Together Event.Back

Brunner, FS, Payne, A, Cairns, E, et al (16 Feb 2023) 'Wastewater genomic surveillance tracks the spread of the SARS-CoV-2 Omicron variant across England', medRxiv.Back

COG-UK (17 Nov 2020) The COG-UK Project Hospital-Onset COVID-19 Infections (HOCI) Study.Back

COG-UK (25 Jan 2021) 'What can wastewater tell us about COVID-19?', Blog.Back

COG-UK Research Project Fund (n.d.) 'Expanding the COG-UK research portfolio to increase impact from genome data'.Back

Kousathanas, A, Pairo-Castineira, E, Rawlik, K, et al (7 March 2020) 'Whole-genome sequencing reveals host factors underlying critical COVID-19', Nature.Back

Pairo-Castineira, E, Clohisey, S, Baillie, KJ, et al, (11 Dec 2020) 'Genetic mechanisms of critical illness in COVID-19', Nature, 591, 92-98.Back

Queen Mary London (7 March 2022) 'Genetic study gives extensive insights into severe Covid-19'.Back

Sanger Institute (14 Oct 2021) 'The inside story of England COVID pandemic described in new study'.Back

Stirrup, O, Hughes, JH, Parker, M, et al (29 June 2021) 'Rapid feedback on hospital onset SARS-CoV-2 infections combining epidemiological and sequencing data', eLife.Back

Interview transcripts

Interview with James Blackstone, Clinical Project Manager, University College London.Back

Interview with Judith Breuer, Professor of Virology, University College London.Back

Interview with Sir Professor Mark Caulfield, Vice Principal for Health, Queen Mary University’s Faculty of Medicine and Dentistry, COG-UK contributor, former Chief Scientists at Genomics England.Back

Interview with Sharon Peacock, Professor of Public Health and Microbiology in the Department of Medicine, Cambridge University and Executive Director of the COVID-19 Genomics UK (COG-UK) Consortium.Back