Ag Plenus launches new AI model to predict antifungal activity
AgPlenus announced the launch of its Antifungal Potency Prediction Model (APP), a machine learning model that predicts the antifungal potency of small molecules directly from their chemical structure, expanding the capabilities of Evogene's ChemPass AI for Ag™ platform to assess biological potency prior to chemical synthesis and fungal assay validation. The global fungicide market size is approximately US$22 billion annually. Fungal diseases cause serious crop losses worldwide and generate tens of billions of dollars in economic losses every year, posing an increasingly serious threat to food security. At the same time, widespread and repeated use of existing fungicides has accelerated the emergence of resistant fungal pathogens and reduced the long-term effectiveness of many commercial products. The APP model uses advanced machine learning algorithms and is trained based on AgPlenus' proprietary data set. The model is capable of predicting antifungal potency in the early stages of discovery, supporting decision-making prior to chemical synthesis and biological testing, significantly reducing the number of molecules that require experimental evaluation and focusing resources on candidates most likely to achieve downstream development success. The models launched are expected to support and advance the AgPlenus internal fungicide product pipeline, including the APTF-1 target against devastating global crop diseases such as septicarum leaf blight in wheat, and contribute to pipeline expansion against other key pathogens such as Botrytis and Fusarium. AgPlenus CEO Dan J. "By enabling us to predict antifungal potency directly from molecular structure prior to chemical synthesis, the APP model allows us to identify and prioritize high-quality candidate molecules at the earliest stages of discovery," said Dr. Gelvan. The model also provides the basis for other predictive AI models that AgPlenus and Evogene plan to jointly develop to predict other key biological attributes during crop protection discovery.