Development and application of an advanced algorithm for environmental risk assessment and safety management of pesticide residues in agricultural soils: Monitoring of currently used pesticide in upland soils.

This study conducted the development of an advanced risk assessment algorithm system and safety management strategies using pesticide residue monitoring data from soils. To understand the status of pesticide residues in agricultural soils, monitoring was performed on 116 types of pesticides currently in use across 300 soil sites. The analysis of the monitoring results, alongside the physicochemical properties of the pesticides, led to the selection of soil half-life as a critical component in residue analysis. The use of Toxicity Exposure Ratio (TER) and Risk Quotient (RQ) for environmental risk assessment, based on monitoring data, presents limitations due to its single-component, conservative approach, which does not align with actual field conditions. Therefore, there is a necessity for a risk assessment process applicable in real-world scenarios. In this research, an efficient and accurate risk assessment algorithm system, along with a safety management model, was developed. Using the physicochemical properties of pesticides (such as soil half-life), monitoring results, and toxicity data, cluster analysis and Principal Component Analysis (PCA) validation identified four pesticides: boscalid, difenoconazole, fluquinconazole, and tebuconazole. The k-mean cluster analysis selected three priority management sites where the contribution of these four pesticides to the RQ was between 94-99 %, showing similar results to the RQ calculated for all pesticides. Predictions made with the developed model for the time required for soil half-life based RQ to drop below 1 at these priority sites showed only a 1-9 day difference between the four pesticides of concern and all pesticides, indicating comparable outcomes. The scenario of replacing high-risk pesticides with those of lower risk demonstrated that the RQ could be consistently maintained at about 50 % level. The results of this study suggest that through monitoring, evaluation, and management, effective and accurate environmental safety management of pesticides in soil can be achieved.

Opportunities and challenges of machine learning in bioprocesses: Categorization from different perspectives and future direction.

Recent advances in machine learning (ML) have revolutionized an extensive range of research and industry fields by successfully addressing intricate problems that cannot be resolved with conventional approaches. However, low interpretability and incompatibility make it challenging to apply ML to complicated bioprocesses, which rely on the delicate metabolic interplay among living cells. This overview attempts to delineate ML applications to bioprocess from different perspectives, and their inherent limitations (i.e., uncertainties in prediction) were then discussed with unique attempts to supplement the ML models. A clear classification can be made depending on the purpose of the ML (supervised vs unsupervised) per application, as well as on their system boundaries (engineered vs natural). Although a limited number of hybrid approaches with meaningful outcomes (e.g., improved accuracy) are available, there is still a need to further enhance the interpretability, compatibility, and user-friendliness of ML models.