Researchers in Graz develop technology for precise assessment of the danger of virus variants
One of the greatest difficulties in combating viral infectious diseases is the excellent adaptability of the viruses. In the case of SARS-CoV-2, we have seen how quickly new variants are constantly being formed that bring with them different properties. That is why it is important that in the future we succeed in predicting more quickly and precisely how dangerous a virus can become," explains Christian Gruber, CEO of the bioinformatics start-up Innophore, a spin-off of the University of Graz and the Austrian Centre of Industrial Biotechnology (acib).
The scientist, in collaboration with colleagues from acib and the University of Graz, is now presenting a new artificial intelligence-based technology that reliably classifies the dangerousness of virus variants. To this end, global and local data, such as from wastewater samples that are quickly and comprehensively tested for excreted viruses, are incorporated into the calculations. The results of the collaboration, which draws on the simulation power of Amazon Web Services, were recently published in the journal "Scientific Reports" from the publication house Nature.
Previous classification systems of virus variants required a lot of time. Christian Gruber and his team now use the point cloud technology developed by Innophore for this purpose. In the first step, structural models of the most current virus variants are created. From these models, the researchers calculate so-called point clouds. "These give us information about how strongly the SARS-CoV-2 spike receptor binding domain (RBD) interacts with the human receptor hACE2. This in turn is an important indicator of infectivity," explains Christian Gruber.
A key advantage of point cloud technology is also that the method is based on artificial intelligence and trained with large data sets. This increases the accuracy and improves the reliability of the predictions. The incorporation of simulation power from Amazon Web Services additionally increases the speed of the calculations performed by Innophore many times over. The collaboration between the two companies was established as part of the Diagnostics Development Initiative (DDI) and has grown into a stable cooperation.
Publication: Gruber et.al, Structural bioinformatics analysis of SARS-CoV-2 variants reveals higher hACE2 receptor binding affinity for Omicron B.1.1.529 spike RBD compared to wild type reference. Scientific Reports volume 12, Article number: 14534 (2022). https://doi.org/10.1038/s41598’022 -18507-y.