AI Predicts Antibiotic Resistance Spread Among Bacteria, Aiding Public Health

Researchers at Chalmers University of Technology have developed an AI model capable of predicting when bacteria will develop resistance to antibiotics. This model, trained on extensive datasets, indicates that resistance spreads more readily among genetically similar bacteria, particularly in environments like humans and wastewater treatment plants. Antibiotic resistance, identified by the World Health Organization (WHO) as a major global health threat, occurs when bacteria evolve to withstand antibiotics, rendering infections harder to treat. The AI model analyzes historical gene transfers between bacteria, utilizing data on their DNA, structure, and environment. It was trained using nearly one million bacterial genome sequences. The study revealed that bacteria in humans and wastewater treatment plants are more prone to developing resistance through gene transfer, due to the high concentration of resistance genes and frequent exposure to antibiotics. The model accurately predicted resistance gene transfer in four out of five cases, suggesting potential for even more precise future models. Researchers aim to use the AI to detect the spread of new resistance genes to disease-causing bacteria and to develop practical solutions, such as improved diagnostics and environmental monitoring. This research, published in Nature Communications, highlights AI's potential in combating antibiotic resistance and protecting public health.