Agemica: Extract from Sunday Times, 12 Jan 202428 May 2025 17:10
very practical approach, in my opinion:
nag, who has started several companies and is an adjunct professor of genetics at stanford university, is not trying to create his vaccine from whole cloth. agemica has created a platform that utilises machine learning — systems that draw out patterns from vast oceans of data — to collate databases of compounds and genomes to suggest novel ****tails of already approved drugs.
focusing on existing medications, rather than inventing new ones, could dramatically reduce the time taken to get a medicine on the market. novel treatments typically take at least a decade to go from discovery to approval.
and indeed, some people are already taking medications off-label, such as the diabetes drug metformin, which has been correlated to longer life but has not been shown to definitively provide life-extension efficacy in a clinical trial.
mixing up old drugs in new ways may seem a longshot, but nag is not alone in his conviction that the dawn of artificial intelligence portends a new era of discovery.
that is because these systems can model complex biological systems in ways previously not possible, invent new compounds and predict their likelihood of success. the potential to shorten timescales, and slash the billions of dollars required to get a single drug to market, is profound. jensen huang, the billionaire founder of nvidia, which makes the chips that power the world’s top ai models, said in a recent speech: “for the very first time, biology has the opportunity to become engineering, not science. when something becomes engineering, it becomes less sporadic and exponentially improves.”
Sign In
Follow the stocks
