Atom-Driven and Knowledge-Based Hydrolysis Metabolite Assessment for Environmental Organic Chemicals

Area of Science:

• Environmental Chemistry
• Computational Chemistry
• Biotechnology

Background:

• Environmental organic chemical metabolism is crucial for assessing ecological impact and safety.
• Hydrolysis is a key metabolic and degradation pathway for these chemicals.
• Traditional experimental methods for identifying hydrolysis products are time-consuming and expensive.

Purpose of the Study:

• To develop and validate machine-learning-based atomic-driven models for predicting hydrolysis reactions in environmental organic chemicals.
• To integrate these models with an expert system for a global hydrolysis prediction approach.
• To enhance the efficiency and accuracy of identifying hydrolysis metabolites.

Main Methods:

• Construction of machine-learning-based atomic-driven models to predict hydrolysis reactions.
• Categorization of hydrolysis into four main sites: N-Hydrolysis, O-Hydrolysis, C-Hydrolysis, and Global-Hydrolysis.
• Integration of machine learning models with a knowledge-based expert system to prioritize hydrolysis products.

Main Results:

• The global hydrolysis site prediction model achieved 93% accuracy on an external test set of 75 chemicals.
• The model successfully predicted 90 out of 99 experimental hydrolysis products, achieving a 90% hit rate.
• The developed model demonstrates high efficacy in identifying potential hydrolysis metabolites.

Conclusions:

• Machine-learning models offer a powerful, efficient alternative to traditional methods for predicting chemical hydrolysis.
• The integrated global hydrolysis model significantly improves the identification of environmental chemical metabolites.
• This approach holds substantial potential for environmental risk assessment and chemical safety evaluations.