Data-Driven Approach for Designing Eco-Friendly Heterocyclic Compounds for the Soil Microbiome.

Soil microbiota plays crucial roles in maintaining the health, productivity, and nutrient cycling of terrestrial ecosystems. The persistence and prevalence of heterocyclic compounds in soil pose significant risks to soil health. However, understanding the links between heterocyclic compounds and microbial responses remains challenging due to the complexity of microbial communities and their various chemical structures. This study developed a machine-learning approach that integrates the properties of chemical structures with the diversity of soil bacteria and functions to predict the impact of heterocyclic compounds on the microbial community and improve the design of eco-friendly heterocyclic compounds. We screened the key chemical structures of heterocyclic compounds─particularly those with topological polar surface areas (<74.2 Å or 111.3-154.1 Å), carboxyl groups, and dissociation constant, which maintained high soil bacterial diversity and functions, revealing threshold effects where specific structural parameters dictated microbial responses. These eco-friendly compounds stabilize communities and increase beneficial carbon and nitrogen cycle functions. By applying these design parameters, we quantitatively assessed the eco-friendliness scores of 811 heterocyclic compounds, providing a robust foundation for guiding future applications. Our study disentangles the critical chemical structure-related properties that influence the soil microbial community and establishes a computational framework for designing eco-friendly compounds with ecological benefits from an ecological perspective.