AI-Designed Synthetic Polymers Overcome Superbug Resistance

AI-designed synthetic polymers have been shown to physically destroy antibiotic-resistant bacteria, bypassing superbug resistance.

According to a report by Azure Life Science on the 25th (local time), researchers at Stanford University’s School of Engineering developed an AI model to identify polymers that mimic the antimicrobial action of natural peptides.

The researchers generated 1.7 million polymer candidates and used AI to predict their antimicrobial properties. Because data on antimicrobial peptides are abundant but polymer data are limited, they first trained the model on the chemical properties of peptides and then applied what it learned to polymers.

Eric Appel, a professor of materials science at Stanford, explained, “These molecules attack bacteria through a physical mechanism, not a biochemical one. It is much harder for bacteria to change the lipid structure or surface charge of their entire cell membrane than to evolve to reject chemical drugs.”

The researchers selected 20 candidates on which the models disagreed most, tested them and fed the results back into the model to improve its performance. They then synthesized the 10 final candidates and evaluated their antimicrobial activity against E. coli.

All 10 candidates performed better than expected, and one showed strong efficacy against biofilms, which are difficult to treat with existing antibiotics.

Shoshana Williams, a former graduate researcher at Stanford, said, “These 10 candidates rank among the most potent antimicrobial polymers reported to date.”

The researchers said the technology is effective against gram-negative bacteria such as E. coli as well as gram-positive bacteria such as Staphylococcus aureus. Williams noted, “No new class of antibiotics targeting gram-negative bacteria has emerged in decades.”

Professor Appel said, “It may be possible to create tailored antimicrobial polymers that eliminate only the specific strain causing an infection while preserving human tissue and beneficial bacteria. This process opens a promising path to finding new antibiotics to treat serious and complex infections.”

The findings were published in the international journal Matter.