Artificial intelligence aids in creating antibiotics with novel bacterial assault methods

In a groundbreaking development, a team of researchers, including Neil Withers, Philip Robinson, Jamie Durrani, Julia Robinson, and Chris Pink, have made significant strides in the fight against antibiotic-resistant bacteria. The team, based at the Massachusetts Institute of Technology (MIT), has used artificial intelligence to develop two new potent antibiotics, named NG1 and halicin, effective against drug-resistant gonorrhea and MRSA bacteria.

The urgent public health crisis due to antibiotic resistance has led to the exploration of non-antibiotics with antibacterial activity. The development of new antibiotics has failed to keep pace with rising antibiotic resistance in bacteria, resulting in approximately five million deaths each year.

The team used a graph neural network to sift through over 45 million chemical fragments for potential antibacterial activity against Neisseria gonorrhoeae and Staphylococcus aureus. After filtering out undesirable properties, the AI models designed over 36 million compounds and screened them for antibacterial activity.

Two molecules, NG1 and DN1, stood out for their potency, selectivity, and new mechanisms of action against drug-resistant bacteria. NG1 selectively targets pathogenic N. gonorrhoeae while sparing beneficial commensal species, and kills a highly drug-resistant strain found in the US. DN1, on the other hand, shows strong bactericidal activity against S. aureus with potency comparable to frontline antibiotics like vancomycin and linezolid. It achieves complete killing of S. aureus within four hours, while vancomycin requires around ten hours.

Jonathan Stokes, a computational chemist at McMaster University in Canada, calls the discovery an 'interesting demonstration' of generative AI's use in early-phase antibiotic discovery. However, he highlights two limitations of AI-aided approaches: many AI-generated structures remain impossible to synthesize, and the mechanism of action of AI-designed antibiotics needs to be unraveled.

A feature titled "Charting the rise in antimicrobial resistance" was published on 2023-09-04, and an article titled "Explainer: A new antibiotic class has been discovered - how excited should we get?" was written by Julia Robinson. A story titled "Research" was published by Kira Welter, but the date is unspecified.

This discovery marks a significant step forward in the battle against antibiotic-resistant bacteria, offering a glimmer of hope in the ongoing public health crisis. As research continues, the potential of AI in antibiotic discovery is becoming increasingly clear.

Artificial intelligence aids in creating antibiotics with novel bacterial assault methods