“Putting data in context – that’s the beauty of it”
Interview: Jacques Schrenzel on the current state of the key technology Whole Genome Sequencing (WGS)and what it can do to combat antibiotic resistance.
On 17 August, the "One Health Meets Sequencing" Symposium 2021 took place as an online event. It presented current and ongoing projects in One Health and Whole Genome Sequencing in Switzerland, complemented by international experts discussing the latest scientific advances in the field. This year's focus was on antibiotic resistance, linking scientific insights to practice, not least to questions about surveillance and outbreak management.
Since whole genome sequencing (WGS) captures all resistance genes, including those acquired by horizontal gene transfer between different bacterial species, the technology is ideal for tracing antibiotic resistance across the entire biological system comprising humans, animals and the environment (One Health approach). Thus, it has become a key technology in the study of the emergence and spread of antibiotic resistance. In this interview, Jacques Schrenzel, Head of the Genomic Research and Bacteriology Laboratories of the University Hospitals of Geneva (HUG) and co-organiser of the symposium, talks about where the field currently stands, what visions it is pursuing and what it can contribute to the fight against antibiotic resistance.
After 2019, the symposium was held for the second time this year. What developments have taken place in the field of WGS within these two years, Professor Schrenzel?
It needs to be emphasised that two years ago we were still talking a lot about the difficulties of generating and assembling a genome. Two years later, most technical steps are described much more clearly or even automated, the WGS technology is more widely disseminated. Many of this year's excellent presentations underlined how potent it has become. The focus has now shifted to the questions we want to address and how we can make the most out of the data generated by WGS.
Could you explain that in more detail?
Joachim Frey gave a good example at the symposium: he demonstrated the crucial role WGS played in clarifying the worldwide spread of anthrax. He showed that besides the technology itself, obtaining metadata on each sample from all over the world was essential. These metadata refer to when a sample was taken, where, on which patient, which animal, etc. In the case of anthrax, this revealed that the known human infections in Switzerland were all in employees of wool-processing companies. Only on the basis of this information did the genetic information about relationships between different samples begin to make sense and gradually an accurate picture emerged of how anthrax had spread worldwide through the international trade in goat and sheep wool. That's the beauty of it: if we put pure sequencing data in context, we can answer important questions with WGS.
What then are the big questions that WGS can help to resolve in the AMR field?
One of them is: how is antibiotic resistance selected in the environment? Jesse Shapiro presented an elegant project on this during the symposium. It became clear that WGS has the potential to clarify narrowly defined questions, but that really requires intelligent study designs. On the other hand, there is also the potential to capture the big picture, so to speak - to clarify which proportion of resistances can actually be traced back to which specific activities. For example, which resistances arise from the use of antimicrobials in animal breeding and where are they then found again in clinically relevant human pathogens? By being able to record and compare all resistance genes using whole genome sequencing, we can identify connections across all sectors and map them geographically without limits. Again, assuming we get good data that also provides context through metadata.
In order to accurately capture and quantify the multiple routes of antibiotic resistance spread, huge amounts of data would need to be compiled, I assume?
This again underlines the need to have interoperable datasets and to standardise metadata. Thus, a lot of discussions this year focused more on the question of how we can bring together the large amounts of data on the genomes of antibiotic-resistant pathogens that are currently being generated in human and veterinary medicine, in environmental laboratories, but also in many research projects. The more data that comes together and can be immediately compared, because it is interoperable, the more revealing it is.
How can this lead to tangible public health benefits?
At present, we suspect several activities to be the main drivers of antibiotic resistance development, but we do not in fact know their actual contribution. We need to quantify the various elements contributing to the spread of antibiotic resistance. This would show us exactly where we need to act and we would then have a much clearer idea of what effect which intervention would have. This would also enable policy-makers to take bold and targeted action. That is why sharing WGS data from all sectors is indeed of the utmost importance for public health.
Should laboratories that produce WGS data be obliged to share them via a common database?
I think a blunt top-down approach would not be ideal. But the government could certainly facilitate data sharing by supporting an appropriate database for the long term. Researchers and institutions should be able to feed their data into it with ease. This means that standards must first be defined, especially for metadata, and then professional support must be offered to all laboratories, for example by sharing tools to facilitate the upload of standardised data and metadata to the common database.
At least SNSF-funded researchers already have to deposit data generated in their work in repositories.
Unfortunately, while deposited research data is technically accessible, it remains practically useless. Because there are no clear standards on interoperability. That's exactly where good support for researchers could be of great benefit. They have to deposit the data anyway. And with the platform for analysing WGS data developed by the Swiss Institute of Bioinformatics as part of an NRP 72 project, this aspect is also already at hand. It doesn't really take that much more going forward.
So what stage might the discussion have reached by the time the next symposium is held, which is already next year?
Hopefully we will already be talking about international infrastructures and interoperability. Antibiotic resistance is after all spreading worldwide and prompting us to develop rapidly, to the same scale.