This could re-energise the drug pipeline just in time
Could AI save us from the antibiotic apocalypse?
That’s the hope from new research in which scientists designed novel drugs atom-by-atom to beat two notorious superbugs, including MRSA.
Antibiotic resistance is a growing problem, causing the deaths of around five million people a year.
Even routine operations could become life-threatening events in future – unless a new generation of antibiotics can be found.
Enter the data crunching power of artificial intelligence.
Researchers at the Massachusetts Institute of Technology used generative AI to design millions of possible compounds that don’t exist or haven’t yet been discovered by science – and then screened them for likely activity against bacteria, as well as toxicity in human cells.
The top candidates were structurally unlike any existing antibiotics and appeared to work in completely novel ways to disrupt the membranes of bacterial cells.
Tests on bacteria in lab dishes and infected mice identified two possible antibiotics against MRSA and the bacteria that causes gonorrhoea.
There would be many years of further testing, including clinical trials, before the drugs would be prescribed by doctors.
But the new approach could prove significant.
The last major class of antibiotics was discovered in the 1980s.
There have been tweaks to the drugs since, but that only slowed the development of resistance.
Many pharmaceutical companies have given up on antibiotic development, daunted by the challenge and the cost.
Using AI could take out some of the risk of failure. And that might re-energise the drug pipeline just in time.
Source link
