Taxonomy-agnostic hyperspectral-morphological phenotyping of fungal pathogen chemical-stress responses using machine learning
This study demonstrates that a taxonomy-agnostic workflow integrating hyperspectral imaging, quantitative morphology, and machine learning can accurately predict the crop-of-isolation (coffee vs. cacao) of *Colletotrichum* fungal isolates based on their standardized chemical-stress response fingerprints, offering a rapid, DNA-free method for high-throughput antifungal screening.