BioNTech and InstaDeep collaboration develops SARS-CoV-2 Early Warning System

In a recent announcement, BioNTech and InstaDeep revealed the companies’ developed Early Warning System (EWS), a new computational method designed to analyse worldwide sequencing data and able to predict high-risk variants of SARS-CoV-2.

The collaboration between BioNTech and InstaDeep is modelled on artificial intelligence (AI) calculated immune escape and fitness metrics.

AI algorithms are combined with the structural basis of the viral spike protein to alert the presence of potential high-risk variants entered into the SARS-CoV-2 sequence data archives in less than a day. The data is based on measurements scoring fitness – eg ACE2 and variant spike protein interaction – alongside immune escape properties.

The predictions have been supported by evidence using experimental data produced through both in-house facilities set up in London and from data available to the public.

Speaking on the potential capabilities of the EWS, Ugur Sahin, chief executive officer and co-founder of BioNTech, said: “Early flagging of potential high-risk variants could be an effective tool to alert researchers, vaccine developers, health authorities and policy makers, thereby providing more time to respond to new variants of concern.”

BioNTech confirmed that a paper written on the new AI developments has been made available on BioRxiv – a pre-print server – and has since been submitted to a peer-reviewed journal.

Researchers involved in the AI study shared that the continuing COVID-19 pandemic ‘is leading to the discovery of hundreds of novel SARS-CoV-2 variants on a daily basis’. They added ‘while most variants do not impact the course of the pandemic, some variants pose significantly increased risk when the acquired mutations allow better evasion of antibody neutralisation in previously infected or vaccinated subjects, or increased transmissibility.’

In its trial period, the EWS identified more than 90% of the World Health Organization (WHO) designated variants – Alpha, Beta, Gamma, Theta, Eta and Omicron – on an average of two months in advance. The same variants were detected by the system in the same week its sequence was initially uploaded, with the Omicron variant ranked in the high-risk category on the same day that its sequence became available.

Rather than using a ‘wait-and-watch’ approach, the EWS ranks SARS-CoV-2 variants based entirely on existing data, eliminating the need for further data to trickle through.

On the same day of EWS detection, the Omicron variant was first uploaded as the highest immune escaping variant from over more than 70,000 recorded variants discovered between early October 2021 and late November 2021.