Learning the language of plant immunity: opportunities and challenges for AI-assisted modelling of fungal effector x...

Phytopathogenic fungi cause substantial crop losses worldwide. They secrete proteins called effectors, which enable infection through interactions with diverse host proteins. These interactions are fundamentally important to plant disease and its practical control. New artificial intelligence (AI) techniques can predict many individual protein structures to near experimental accuracy. Although these techniques also predict protein complexes, they are not as accurate as single-protein models. Use of AI to study interactions between fungal pathogen effectors and plant proteins is currently limited. However, despite some challenges, early adoption of AI has highlighted its potential. General improvements in AI-assisted protein complex modelling may create more opportunities in future. This review focuses on recent research using AI to study the interactions between fungal effectors and plant proteins, outlining challenges and emerging opportunities.