AI Predicts Antibiotic Resistance Spread in Bacteria via Gene Transfer

An AI model, trained on extensive genetic data, can now predict antibiotic resistance development in bacteria. A study from Chalmers University of Technology indicates that antibiotic resistance is more readily transmitted between genetically similar bacteria, particularly in wastewater treatment plants and within the human body. Erik Kristiansson, Professor at Chalmers University of Technology and the University of Gothenburg, states, "Bacteria that are harmful to humans have accumulated many resistance genes... Our research examines this complex evolutionary process to learn how these genes are transferred to pathogenic bacteria. This makes predicting how future bacteria develop resistance possible." The AI model, detailed in *Nature Communications*, analysed historical gene transfers using bacterial DNA, structure, and habitat data. The model was trained on nearly a million bacterial genomes. David Lund, doctoral student at Chalmers and the University of Gothenburg, notes, "We see that bacteria found in humans and water treatment plants have a higher probability of becoming resistant through gene transfer… Another important factor that increases the likelihood that resistance genes will 'jump' from one bacterium to another is the genetic similarity of the bacteria." The model's accuracy was tested against known resistance gene transfers. In four out of five cases, the model accurately predicted the transfer. Researchers hope to refine the model for use in diagnostics and monitoring, potentially improving the detection of multi-resistant bacteria and monitoring environments where antibiotics are present.