Machine learning for designing low-risk microbial consortia pesticides.
Machine learning for designing low-risk microbial consortia pesticides.
Mónica Garcés-Ruiz,Belén Guijarro Díaz‐Otero,3 Authors,Stéphane Declerck
TLDR
The methods and steps necessary to establish reliable SynComs are reviewed and how AI and ML approaches could improve the construction and validation of SynCom-based LRPs to obtain more specific results that can contribute to their risk assessment are described.
Abstract
Microbial consortia, considered low-risk pesticides (LRPs), appear to be valuable tools for reducing our dependence on chemical pesticides. However, their use is limited by inconsistent product efficacy and registration difficulties. Artificial intelligence (AI) and machine learning (ML) offer solutions for designing and evaluating synthetic microbial communities (SynComs), predicting their compatibility, ecological stability, and biocontrol efficacy. The transition from laboratory discovery of SynCom-based LRPs to field application and commercialization could be significantly accelerated. Here, we review the methods and steps necessary to establish reliable SynComs and describe how AI and ML approaches could improve the construction and validation of SynCom-based LRPs to obtain more specific results that can contribute to their risk assessment.