Construction of an antidepressant priority list based on functional, environmental, and health risks using an interpretable mixup-transformer deep learning model.

As emerging pollutants, antidepressants (AD) must be urgently investigated for risk identification and assessment. This study constructed a comprehensive-effect risk-priority screening system (ADRank) for ADs by characterizing AD functionality, occurrence, persistence, bioaccumulation and toxicity based on the integrated assignment method. A classification model for ADs was constructed using an improved mixup-transformer deep learning method, and its classification accuracy was compared with those of other models. The accuracy of the proposed model improved by up to 23.25 % compared with the random forest model, and the reliability was 80 % more than that of the TOPSIS method. A priority screening candidate list was proposed to screen 33 high-priority ADs. Finally, SHapley Additive explanation (SHAP) visualization, molecular dynamics, and amino acid analysis were performed to analyze the correlation between AD structure and toxic receptor binding characteristics and reveal the differences in AD risk priority. ADs with more intramolecular hydrogen bonds, higher hydrophobicity, and electronegativity had a more significant risk. Van der Waals and electrostatic interactions were the primary influencing factors, and significant differences in the types and proportions of the main amino acids in the interaction between ADs and receptors were observed. The results of the study provide constructive schemes and insights for AD priority screening and risk management.

Theoretical design and process control of neonicotinoids insecticides suitable for synergistic degradation with the rubisco enzyme from rhizobia and carbon-fixing bacteria in soil.

In this study, we studied and developed the modification schemes of environmentally friendly substitutes of neonicotinoid insecticides (NNIs) along with the regulatory measures that effectively enhanced the synergistic degradation of NNIs by soil rhizobia and carbon-fixing bacteria. Firstly, the binding ability of NNIs to the two key proteins was characterized by molecular docking; secondly, the mean square deviation decision method, which is a comprehensive evaluation method, was used to investigate the binding ability of NNI molecules with the two Rubisco rate-limiting enzymes. The three-dimensional quantitative structure-activity relationship (3D-QSAR) model was established for the synergistic degradation and single effect of rhizobia and carbon-fixing bacteria. Finally, after combining the 3D-QSAR model with a contour map analysis of the synergistic degradation effect of soil rhizobia and carbon-fixing bacteria, 102 NNI derivatives were designed. Flonicamid-36 and other four NNI derivatives passed the functional and environmentally friendly evaluation. Taguchi orthogonal experiment and factorial experiment-assisted molecular dynamics method were used to simulate the effects of 32 regulation schemes on the synergistic degradation of NNIS and its derivatives by rhizobia and carbon fixing bacteria. The synergistic degradation capacity of soil rhizobia and carbon-fixing bacteria was increased to 33.32% after right nitrogen supplementation. This indicated that supplementing the correct amount of nitrogen in the soil environment was beneficial to the microbial degradation of NNIs and their derivatives.