Advances in deep learning for antimicrobial research

The rapid spread of antimicrobial resistance and the lag in the development of new antimicrobial drugs have become major challenges for global public health. Conventional methods for discovering antimicrobial drugs are characterized by long development cycles and high costs, which are further exacerbated by the rapid evolution of bacterial resistance. Therefore, there is an urgent need to develop efficient computational methods to accelerate the identification of drug resistance mechanisms and the research and development process of new antimicrobial drugs. In recent years, the new generation of artificial intelligence technologies represented by deep learning (DL) has made remarkable progress in antibacterial research. This paper reviews the current application status of DL in antimicrobial research, aiming to provide a comprehensive overview that covers background knowledge including technologies and principles, related case studies, and future perspectives. Firstly, the development of antibacterial drug discovery, the advancement of sequencing technique and basic concepts related to DL are introduced. Subsequently, a critical evaluation is conducted on the representative reports of DL in predicting drug resistance mechanisms and promoting antimicrobial drug discovery. Finally, the challenges faced by DL in antibacterial research and its future development directions are discussed, aiming to provide constructive suggestions and guidance for researchers in this field.