AI-designed antibiotics pave way for defeating superbugs
10 days ago
- #AI
- #Antibiotics
- #Drug-Resistance
- AI has designed two new potential antibiotics targeting drug-resistant gonorrhoea and MRSA.
- The antibiotics were developed atom-by-atom by AI and showed effectiveness in lab and animal tests.
- These compounds require years of refinement and clinical trials before they can be prescribed.
- MIT researchers believe AI could usher in a 'second golden age' of antibiotic discovery.
- Drug-resistant infections cause over a million deaths annually, highlighting the urgent need for new antibiotics.
- The study used generative AI to design antibiotics from scratch, evaluating 36 million compounds.
- AI was trained on chemical structures and their effects on bacteria to predict effective new drugs.
- Two approaches were used: one built from chemical fragments, the other gave AI complete freedom.
- The design process filtered out toxic or ineffective compounds and avoided similarities to existing antibiotics.
- Two promising drugs were synthesized and tested, but further refinement is needed before human trials.
- Experts highlight the potential of AI in drug discovery but stress the need for rigorous safety and efficacy testing.
- Economic challenges exist, as new antibiotics must be used sparingly, reducing profitability for developers.