Antibiotics play an important role in the treatment of bacterial infection, but it is a "double-edged sword", which also promotes bacteria to strengthen drug resistance. An international team recently reported in the American journal Science that they have developed an antibiotic prescription algorithm using machine learning technology and genome sequencing technology, which can reduce the risk of drug resistance in infection treatment by half. The focus of treating infection is to correctly match the drug resistance of antibiotics and pathogens. However, even if the match is correct, antibiotic resistance may still occur. One reason is that bacteria may develop drug resistance by random mutation in evolution, but this random process is difficult to predict and avoid. The research team led by the Israeli Institute of technology found that the drug resistance of most infected patients was not caused by random mutations of pathogenic bacteria, but by the rapid reinfection of another strain in the patient's microbiome resistant to prescription antibiotics. The researchers turned the findings into a therapeutic idea: the antibiotics used for treatment should not only match the drug resistance of the current pathogen, but also match other bacteria in the patient's microbiome that may replace the current pathogen. Using the microbiome data records of more than 200000 patients for 8 years, the research team constructed a machine learning algorithm model to predict the risk of individual resistance to specific antibiotics. The researchers also used a large number of antibiotic prescription data for the treatment of urinary tract infection and wound infection to train the algorithm, so that it can formulate personalized antibiotic treatment prescriptions. Research shows that the antibiotic prescription algorithm can halve the risk of antibiotic resistance in treatment. "We found that the sensitivity of patients to antibiotics in past infections can be used to predict the risk of drug-resistant infections after they receive antibiotics again." Dr. Matthew strasi, the first author of the paper and a researcher at the Israeli Institute of technology, explained. The researchers hope that this study will provide doctors with better tools to develop personalized antibiotic treatment programs to improve efficacy and minimize the spread of drug-resistant pathogens. (Xinhua News Agency)