Machine Learning- and AI-Driven QSAR Models for the Discovery of Novel Potential Fungicides: FAPI (Fungicide Targeting Acid Phosphatase Inhibition)

Abstract

Fungal pathogens like Podosphaera xanthii (powdery mildew) and Botrytis cinerea (gray mold) cause significant agricultural losses, with fungicide resistance escalating due to the overreliance on conventional treatments. Consequently, the development of sustainable alternatives with novel modes of action is imperative for future crop protection. Acid phosphatases (APs), which play a key role in fungal phosphate metabolism and virulence, have emerged as promising molecular targets. To accelerate the identification of fungicides targeting acid phosphatase inhibition (FAPI), machine learning (ML), and artificial intelligence (AI)-driven quantitative structure–activity relationship (QSAR) models incorporating topological molecular descriptors have been employed to predict fungicidal activity. The experimental validation of the predicted candidates highlights the promising potential of these novel fungicides and underscores the value of ML- and AI-based QSAR methodologies in the development of next-generation, resistance-breaking fungicidal agents.