Predicting SARS-CoV-2 Omicron Variant Impact with AI

When the genome of the SARS-CoV-2 Omicron variant (B.1.1.529) was released on November 22, 2021, it caused significant media attention due to the large number of mutations it contains.

These raw data have spurred questions around COVID-19 vaccines and antibody therapy efficacy.

Given that neither the structural information nor the experimentally-derived antibody interaction of this variant are available, researchers turned to predictive computational methods to model the mutated structure of the spike protein's receptor-binding domain and posit potential changes to vaccine efficacy.

In this non-peer-reviewed study published by bioRxiv on December 6, 2021, these researchers predict some structural changes in the receptor-binding domain that may reduce antibody interaction, but no drastic changes that would completely evade existing neutralizing antibodies, and therefore current COVID-19 vaccines.

'While in vitro experiments are needed to validate these predictions, the predicted results here suggest that existing neutralizing antibodies will still bind to the mutated spike protein of the Omicron variant.'

'However, it appears that the affinity of Omicrons's RBD for neutralizing antibodies is reduced compared to the reference RBD structures.'

'Though there are a multitude of mutations in the RBD of Omicron, these mutations do not appear to be causing any large conformational change that would totally evade antibody interaction.'

'Given the public health urgency in understanding the impacts of new SARS-CoV-2 variants requires that we act quicker than is possible in a lab.'

'Thus, in silico predictive tools like AlphaFold2 and HADDOCK are important for quickly understanding the biochemistry of variants and can help us infer the epidemiological implications of the variant.'