BioNTech, AI Startup InstaDeep Develop COVID Warning Tool

BioNTech, COVID, InstaDeep

A German biotech firm and a north African artificial intelligence (AI) startup have developed an early warning tool that can the highest risk COVID-19 variants months in advance, according to a new study.

As the Financial Times reported Wednesday (Jan. 12), the program — developed by AI startup InstaDeep and BioNTech, which helped develop one of the leading COVID-19 vaccines — was able to spot more than 90% of variants of concern, including omicron, an average of two months before the designation by the World Health Organization.

The companies’ research shows the program can evaluate the risks of new variants using their spike proteins inside minutes, and monitor them as they evolve “nearly in real time.” Ugur Sahin, CEO of BioNTech, said the tool will be available for free.

“Early flagging of potential high-risk variants could be an effective tool to alert researchers, vaccine developers, health authorities and policymakers, providing more time to respond to new variants of concern,” said Sahin.

The system was able to red-flag omicron the same day its genetic sequence was uploaded to the Gisaid database in late 2021, identifying it as the most concerning of the 70,000 variants discovered last fall, the study said. That was three days before WHO classified Omicron as a variant of concern.

Read more: WHO: Omicron Could Have ‘Major Impact’ on Pandemic

Last month, WHO Director-General Tedros Adhanom Ghebreyesus said that “certain features of omicron, including its global spread and large number of mutations, suggest it could have a major impact on the course of the pandemic.”

“More than 10,000 novel variant sequences are discovered every week and human experts simply cannot cope with complex data at this scale,” said Karim Beguir, co-founder and CEO of U.K.-based InstaDeep. “For the first time, high-risk variants could be detected on the spot, potentially saving months of precious time.”

The Financial Times noted that the study has not yet been peer reviewed.