The highly mutated Omicron variant of COVID-19 may not completely escape vaccine-induced antibodies, according to a new study by US researchers using Artificial Intelligence tools developed by Google's DeepMind.

For our comprehensive coverage and latest updates on COVID-19 click here.

The genome of the SARS-CoV-2 Omicron variant (B.1.1.529) was released on November 22, and has since caused a flurry of attention due to a large number of mutations in the spike protein and elsewhere on the virus. The mutations raise concerns that this variant will have a considerable escape from vaccine-elicited immunity.

To explore this, a research team from the University of North Carolina at Charlotte, 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. The study, posted on pre-print server and not yet peer-reviewed, showed "some structural changes in the receptor-binding domain that may reduce antibody interaction, but no drastic changes that would completely evade existing neutralising antibodies (and therefore current vaccines)", as per Colby T. Ford, from the Department of Bioinformatics and Genomics at the varsity.

The study was conducted using AlphaFold2 - a neural network-based deep learning model created by Google DeepMind - which can predict protein structure based on the genetic code. Researchers found that existing neutralising antibodies will still bind to the mutated spike protein of the Omicron variant. The mutations "do not appear to be causing any large conformational change that would totally evade antibody interaction," Ford said.

The antibodies elicited from vaccines or a previous infection will provide some protection against Omicron, he noted. Determining the actual structure of a protein is a time-consuming process, and quantifying protein-protein interactions (like spike-to-antibody interactions) are also experimentally difficult to perform in vitro. Thus the study results will need to be confirmed in laboratory experiments, the team said.

"Given the public health urgency in understanding the impacts of new SARS-CoV-2 variants quickly requires that we act quicker than is possible in a lab," Ford said. Meanwhile, US drug maker Pfizer on 8 December announced that a third dose of its mRNA vaccine can protect against the Omicron variant of COVID-19.

Results from an initial laboratory study demonstrated that serum antibodies induced by the Pfizer-BioNTech COVID-19 vaccine (BNT162b2) neutralises the SARS-CoV-2 Omicron variant after three doses, the company said in a statement.

A third dose also strongly increases CD8+ T cell levels against multiple spike proteins which are considered to correlate with the protection against severe disease. Compared to the wild-type virus, the vast majority of these remain unchanged in the Omicron spike variant, the company said.

The findings were also supported by a small study led by South African researchers that showed that only two doses of Pfizer's COVID vaccine may be up to 40 times less effective against the new super mutant Omicron variant compared with the original virus.

According to Prof Alex Sigal, a virologist at the Africa Health Research Institute, who led the research, Omicron's ability to escape vaccine antibodies is "incomplete". He added that "previous infection, followed by vaccination or booster is likely to increase the neutralisation level and likely confer protection from severe disease in Omicron infection".