Genome Detectives project update, April 2024

With our break taking longer than anticipated, you may be wondering what is happening with Genome Detectives. There is currently a lot of work going on behind the scenes, namely making improvements based upon what we have learned so far, and simultaneously developing ideas for next-level ‘genome detecting’.

We are incredibly grateful to all Zooniverse Genome Detective volunteers for your contributions in moving bacterial genome curation forwards. There are more than 4,500 volunteers who have engaged with the project to date, and many of you have shown exceptional commitment to the tasks. The Genome Detectives project is the first of its kind, and we thank you for your patience while we continue to develop the project with your help. Your enthusiasm and engagement with the project has gone beyond our wildest expectations.

Thanks to the results from this project, we have been able to make improvements to our automated curation scripts on the PubMLST database. This has a huge impact on the time we spend on the essential bacterial gene curation tasks that underpin all of the research we do. It has been very important to undertake these steps carefully, since any new alleles we add to the PubMLST database form the basis by which other alleles are found, and a mistake at this stage would prove costly. Thanks to the greater number of bacterial isolates and curated genes now in the database, we have been able to identify genes that are most informative for research, and remove those, which for multiple reasons, prove to be more troublesome.

Whilst the accuracy of computing scripts is ever improving, there still remains a proportion of bacterial genes that require a human eye to work through a ‘puzzle’ determining how and why changes have occurred in a DNA sequence, and if, for example, these are artefacts, or whether a gene will still be functional etc. We hope we have been able to give you a flavour of the sorts of issues genome detectives need to investigate with the ‘Training Academy’ website ( https://genomedetectives.web.ox.ac.uk/ ).

Our next ambition is to provide a follow-on Genome Detectives platform where you will able to undertake these more complex tasks for yourselves. This will greatly increase the power of the ‘human eye’ in identifying new patterns that AI cannot solve, whilst enabling citizen scientists to continue contributing to infectious disease research. Such an ambitious project inevitably requires time, planning and resources to put in place. We are also investigating collaborative options with other research groups to bring you more areas of exciting science.

We plan to retain the original Zooniverse Genome Detectives project – albeit with some small improvements as suggested by volunteers – as a tool for training, education and outreach, where it has already been well received. We very much enjoy meeting those of you who can attend science fairs and events and find this aspect of working in science especially rewarding. We thank you for your continued interest in Genome Detectives.
Image by Wilhelm Gunkel on UnSplash

Project update

We would like to thank you so much for the wonderful response to the Genome Detectives project that was launched in July 2022! In a matter of a few weeks nearly 1,800 of you have participated as volunteers, with many of you making a significant contribution to the 120,000 plus gene classifications that have been completed to date. We most certainly could not have done this without you!

One of the first results was the discovery, BY YOU, of an unusual sequence pattern for Acinetobacter – a bacterium that causes serious problems for hospitilised patients. We are often asked why the curation process cannot be automated. In fact, we already assign 97% of sequencing data using our existing computer algorithms, and it is only the remaining 3% that are uploaded to Genome Detectives. Here is an example of an interesting anomaly that requires further investigation by scientists and simply would not have been picked up by computer algorithms alone. That said, we are very grateful to the volunteers who have taken the time to contact us with suggestions to automate the simpler tasks on Genome Detectives. With everyone’s help we are continually striving to refine the existing project and we have some exciting plans to develop a second Genome Detectives project for those who enjoy more of a challenge! With such a complex subject, designing tasks that are both understandable and enjoyable is a difficult undertaking and we have learnt a lot, thanks again to your help!

Case studies

Antibiotic resistance and food poisoning

Genome-based research is helping us to understand how levels of antimicrobial resistance (AMR) have been increasing in Campylobacter, the predominant bacterial cause of food poisoning (gastroenteritis) in humans. The bacterium becomes resistant to fluoroquinolones, an antibiotic commonly used to treat clinically vulnerable patients, through a single mutation in the gyrA gene – that is a change in just one of the nucleotide bases that make up the DNA of the gene. Other DNA changes can co-exist and increase the level of resistance even further. These can all be detected from genome sequences of the bacterium and show good correlation with more traditional laboratory-based studies, meaning that the efficiency of detection and understanding of underlying mechanisms can be improved. We have also been able to gain insights into what has been driving the dramatic increase in AMR in Campylobacter over the past two decades. Studies from very large collections of carefully sampled Camplyobacter collections have shown us that in this case, stains of fluoroquinolone resistant Campylobacter infecting humans are associated with use of antibiotics in livestock, most commonly poultry. Studies of this type will help to inform us how we may best safeguard the health of both humans and animals in the future, and develop the best strategies for antibiotic stewardship.

Vaccines to prevent meningitis and septicaemia

Genome-based research performed by scientists can be used to help doctors and specialists at the medical frontline to make decisions for patients or the public. The bacteria Neisseria meningitidis or the "meningococcus" can cause meningitis and septicaemia in all parts of the world. This infection can be prevented by a number of different vaccines that have been developed since the 1960's. One type of vaccine, developed in the 21st century, uses proteins to trigger the immune system to make antibodies that should protect you in case of future exposure to the meningococcus (specifically targeting MenB). As you have seen, however, any change in the genetic code can change the protein that is made by the bacteria, so if the meningococcus DNA changes, then these vaccines are not always effective. We have developed genomics-based analysis methods to assess the meningococcal proteins found in these vaccines. We can, therefore, very quickly and easily assess large numbers of meningococci and public health teams from around the world use our tools to do this to make vaccine policy decisions. We have also found ways of translating this information for use by doctors, microbiologists or public health specialists, who can use this genomic-derived information to see how effective each protein-based vaccine may be and which vaccine they need to give to stop an outbreak or cluster of cases.