Joining the dots to tackle COVID-19: a bioinformatics approach

Stay at home, protect the NHS, save lives. Wash your hands for 20 seconds and avoid touching your face. Keep two metres apart at all times. While the public health advice to combat COVID-19 has been undeniably simple, the virus itself remains an enigma.

The global death toll is well in excess of 250,000, with some patients fighting for their lives in intensive care units, yet others seemingly shrug it off with no symptoms to suggest they were ever infected.

Since the SARS-CoV-2 virus began spreading rapidly around the world, scientists have been racing to understand its pathogenicity. Global collaborations and open data sharing have yielded vast amounts of data which the bioscience community is using to find a potential treatment to reduce the mortality associated with COVID-19.

The mounting evidence points to a peculiar feature of SARS-CoV-2. Far from being limited to the respiratory tract, its deadly effects have been shown to act all over the body - from the lungs to the gut and the kidneys - even causing strokes in otherwise healthy, younger patients.

One particularly dangerous effect, which occurs in some patients between 7 and 10 days after infection, is a cytokine storm. The virus essentially causes the immune system to go into overdrive. Struggling to cope, the body can fill the lungs with fluid, cause multiple organ failure, and perhaps even lead to some neurological effect.

If we are to stand any chance of developing effective treatments - and managing this pandemic - we have to find out what is driving the response of our body to SARS-CoV-2.

Dr Tamas Korcsmaros and his group are perfectly positioned in this regard, with over ten years’ experience in developing bioinformatics tools and techniques to understand the complex nature of inflammatory bowel disease (IBD) via the multitude of interactions that occur within and between cells occupying the human gut. This approach allows scientists to look at systems as a whole, rather than their individual parts.

The reams of data collected by the international scientific community detail the genomes and mutations of SARS-CoV-2 variants across different locations; the structure of the viral proteins; their host targets in human cells; the transcriptomics changes in infected cells; cell or tissue-level differences in the blood or in the body of COVID-19 patients; and human genomic information from patients.

“At present, these data are not connected,” explains Dr Korcsmaros. “Using our demonstrated integrative approaches, we can combine them to produce a more holistic understanding of the effect of SARS-CoV-2, point out non-trivial connections, and facilitate the search for relevant therapeutic options.”

Dr Korcsmaros and his group at Earlham Institute and Quadram Institute are joining with Dr Julio Saez-Rodriguez of the University of Heidelberg as part of the international COVID-19 Disease Map effort.

“Our main goal is to understand the systemic effect of SARS-Cov-2 on our body,” says Dr Korcsmaros. “In particular, why in certain patients the viral infection leads to a much more severe, inflammatory phenotype that causes life-threatening over-activation of our own immune system.

“To achieve this, primarily, we are going to create computational models and then we will test some of the predictions in-house or with expert collaborators.”

To draw an interesting parallel, an article published recently by the NPC stated that “Covid-19 means systems thinking is no longer optional.” This was directed at human civilization from a broad perspective, but the lessons apply equally to biology - as indeed they should. To quote the piece, “Systems thinking goes beyond individual actions to connections, causes and consequences.”

This is precisely the attitude Dr Korcsmaros applies in his work on systems approaches to understanding gut health. Active projects tackle the highly infectious and deadly Salmonella, as well as chronic illness, such as Inflammatory Bowel Disease. His group has developed a range of machine learning and bioinformatics tools that help scientists approach these areas using multi-omics and network-based methods.

SalmoNet, for example, is a resource which links information on how genes and metabolic pathways are regulated in Salmonella, and how proteins interact with each other, looking at all of the interactions within and between multiple strains of the bacterium. PhD student Marton Olbei, who works on the platform, explained in a blog that for SalmoNet “understanding and discovering new biological mechanisms or potential therapeutic intervention targets is one of the most compelling uses.”

SignaLink is another resource which allows users to explore signalling pathways and their cross-talk, including transcription factors, regulatory components and enzymes through an onion-like, multi-layered structure. It’s similar to SalmoNet, but this resource is for humans and model systems such as zebrafish and the nematode worm C. elegans. Other network resources developed by the Korcsmaros Group include NavigOmix and the Autophagy Regulatory Network.

The cytokine storm effect is not unique to COVID-19 but occurs in response to a number of similar infections. The team is also, therefore, comparing the data from the current pandemic to that from others such as the coronaviruses SARS and MERS, as well as H5N1 and H7N9 influenza.

In this way, through identifying specific immune responses involved in the cytokine storm - and perhaps shared pathways - it may well be possible to identify therapeutic targets.