Potential Application of Machine-Learning-Based Quantum Chemical Methods in Environmental Chemistry

Migration mechanism of atrazine in the simulated lake icing process at different freezing temperatures based on density function theory

Atrazine causes concern due to its resistant to biodegradation and could be accumulated in aquatic organisms, causing pollution in lakes. This study measured the concentration of atrazine in ice and the water under ice through a simulated icing experiment and calculated the distribution coefficient K to characterize its migration ability in the freezing process. Furthermore, density functional theory (DFT) calculations were employed to expatiate the migration law of atrazine during icing process. According to the results, it could release more energy into the environment when atrazine staying in water phase (-15.077 kcal/mol) than staying in ice phase (-14.388 kcal/mol), therefore it was beneficial for the migration of atrazine from ice to water. This explains that during the freezing process, the concentration of atrazine in the ice was lower than that in the water. Thermodynamic calculations indicated that when the temperature decreases from 268 to 248 K, the internal energy contribution of the compound of atrazine and ice molecule (water cluster) decreases at the same vibrational frequency, resulting in an increase in the free energy difference of the compound from -167.946 to -165.390 kcal/mol. This demonstrated the diminished migratory capacity of atrazine. This study revealed the environmental behavior of atrazine during lake freezing, which was beneficial for the management of atrazine and other pollutants during freezing and environmental protection.

Application of machine learning in the study of development, behavior, nerve, and genotoxicity of zebrafish

Machine learning (ML) as a novel model-based approach has been used in studying aquatic toxicology in the environmental field. Zebrafish, as an ideal model organism in aquatic toxicology research, has been widely used to study the toxic effects of various pollutants. However, toxicity testing on organisms may cause significant harm, consume considerable time and resources, and raise ethical concerns. Therefore, ML is used in related research to reduce animal experiments and assist researchers in conducting toxicological research. Although ML techniques have matured in various fields, research on ML-based aquatic toxicology is still in its infancy due to the lack of comprehensive large-scale toxicity databases for environmental pollutants and model organisms. Therefore, to better understand the recent research progress of ML in studying the development, behavior, nerve, and genotoxicity of zebrafish, this review mainly focuses on using ML modeling to assess and predict the toxic effects of zebrafish exposure to different toxic chemicals. Meanwhile, the opportunities and challenges faced by ML in the field of toxicology were analyzed. Finally, suggestions and perspectives were proposed for the toxicity studies of ML on zebrafish in future applications.

Experimental insights and modeling innovations: Deciphering Fe(VI) oxidation in imidazole ionic liquids through QSAR and RFR

In this investigation, we conducted a detailed analysis of the oxidation of 16 imidazole ionic liquid variants by Fe(VI) under uniform experimental setups, thereby securing a dataset of second-order reaction rate constants (kobs). This methodology ensures superior data consistency and comparability over traditional methods that amalgamate disparate data from varied studies. Utilizing 16 chemical structural parameters obtained via Density Functional Theory (DFT) as descriptors, we developed a Quantitative Structure Activity Relationship (QSAR) model. Through rigorous correlation analysis, Principal Component Analysis (PCA), Multiple Linear Regression (MLR), and Applicability Domain (AD) evaluation, we identified a pronounced negative correlation between the molecular orbital gap energy (Egap) and kobs. MLR analysis further underscored Egap as a pivotal predictive variable, with its lower values indicating heightened oxidative reactivity towards Fe(VI) in the ionic liquids, leading the QSAR model to achieve a predictive accuracy of 0.95. Furthermore, we integrated an advanced machine learning approach - Random Forest Regression (RFR), which adeptly highlighted the critical factors influencing the oxidation efficiency of imidazole ionic liquids by Fe(VI) through elaborate decision trees, feature importance assessment, Recursive Feature Elimination (RFE), and cross-validation strategies. The RFR model demonstrated a remarkable predictive performance of 0.98. Both QSAR and RFR models pinpointed Egap as a key descriptor significantly affecting oxidation efficiency, with the RFR model presenting lower root mean square errors, establishing it as a more reliable predictive tool. The application of the RFR model in this study significantly improved the model's stability and the intuitive display of key influencing factors, introducing promising advanced analytical tools to the field of environmental chemistry.

A new ChatGPT-empowered, easy-to-use machine learning paradigm for environmental science

The quantity and complexity of environmental data show exponential growth in recent years. High-quality big data analysis is critical for performing a sophisticated characterization of the complex network of environmental pollution. Machine learning (ML) has been employed as a powerful tool for decoupling the complexities of environmental big data based on its remarkable fitting ability. Yet, due to the knowledge gap across different subjects, ML concepts and algorithms have not been well-popularized among researchers in environmental sustainability. In this context, we introduce a new research paradigm-"ChatGPT + ML + Environment", providing an unprecedented chance for environmental researchers to reduce the difficulty of using ML models. For instance, each step involved in applying ML models to environmental sustainability, including data preparation, model selection and construction, model training and evaluation, and hyper-parameter optimization, can be easily performed with guidance from ChatGPT. We also discuss the challenges and limitations of using this research paradigm in the field of environmental sustainability. Furthermore, we highlight the importance of "secondary training" for future application of "ChatGPT + ML + Environment".

Henna plant biomass enhanced azo dye removal: Operating performance, microbial community and machine learning modeling

The bio-reduction of azo dyes is significantly dependent on the availability of electron donors and external redox mediators. In this study, the natural henna plant biomass was supplemented to promote the biological reduction of an azo dye of Acid Orange 7 (AO7). Besides, the machine learning (ML) approach was applied to decipher the intricate process of henna-assisted azo dye removal. The experimental results indicated that the hydrolysis and fermentation of henna plant biomass provided both electron donors such as volatile fatty acid (VFA) and redox mediator of lawsone to drive the bio-reduction of AO7 to sulfanilic acid (SA). The high henna dosage selectively enriched certain bacteria, such as Firmicutes phylum, Levilinea and Paludibacter genera, functioning in both the henna fermentation and AO7 reduction processes simultaneously. Among the three tested ML algorithms, eXtreme Gradient Boosting (XGBoost) presented exceptional accuracy and generalization ability in predicting the effluent AO7 concentrations with pH, oxidation-reduction potential (ORP), soluble chemical oxygen demand (SCOD), VFA, lawsone, henna dosage, and cumulative henna as input variables. The validating experiments with tailored optimal operating conditions and henna dosage (pH 7.5, henna dosage of 2 g/L, and cumulative henna of 14 g/L) confirmed that XGBoost was an effective ML model to predict the efficient AO7 removal (91.6%), with a negligible calculating error of 3.95%. Overall, henna plant biomass addition was a cost-effective and robust method to improve the bio-reduction of AO7, which had been demonstrated by long-term operation, ML modeling, and experimental validation.

Prediction of free radical reactions toward organic pollutants with easily accessible molecular descriptors

Machine learning (ML) is becoming an efficient tool for predicting the fate of aquatic contaminants owing to the preponderance of big data. However, whether ML can "learn" the differences in reactivity among different free radicals has not yet been tested. In this work, the effectiveness of combining ML algorithms with molecular fingerprints to predict the reactivity of three free radicals was evaluated. First, a dataset containing 211 organic pollutants and their respective rate constants with the carbonate radical (CO3•-) was used to develop predictive models using both linear regression and ML methods. The use of topological atomic alignment information, in the form of the molecular access system (MACCS) and Morgan Fingerprint, and the electronic structure features (energy levels of the lowest unoccupied and highest occupied molecular orbitals, ELUMO and EHOMO, and the energy gap between ELUMO and EHOMO) gave satisfactory predictive performances (ML model with Random Forest algorithm with MACCS: RMSEtest = 0.787; linear regression model with energy levels: RMSEtest = 0.641). Additionally, the model interpretation correctly described that the key reactivity features for CO3•- were relatively close to those for SO4•- rather than those for •OH. These results suggest that combination of ML algorithms with easily accessible molecular fingerprints would be a powerful tool to accurately predict the radical reactions towards organic compounds.

Investigation of the Binding Fraction of PFAS in Human Plasma and Underlying Mechanisms Based on Machine Learning and Molecular Dynamics Simulation

More than 7000 per- and polyfluorinated alkyl substances (PFAS) have been documented in the U.S. Environmental Protection Agency's CompTox Chemicals database. These PFAS can be used in a broad range of industrial and consumer applications but may pose potential environmental issues and health risks. However, little is known about emerging PFAS bioaccumulation to assess their chemical safety. This study focuses specifically on the large and high-quality data set of fluorochemicals from the related environmental and pharmaceutical chemicals databases, and machine learning (ML) models were developed for the classification prediction of the unbound fraction of compounds in plasma. A comprehensive evaluation of the ML models shows that the best blending model yields an accuracy of 0.901 for the test set. The predictions suggest that most PFAS (∼92%) have a high binding fraction in plasma. Introduction of alkaline amino groups is likely to reduce the binding affinities of PFAS with plasma proteins. Molecular dynamics simulations indicate a clear distinction between the high and low binding fractions of PFAS. These computational workflows can be used to predict the bioaccumulation of emerging PFAS and are also helpful for the molecular design of PFAS to prevent the release of high-bioaccumulation compounds into the environment.

Contribution of molecular structures and quantum chemistry technique to root concentration factor: An innovative application of interpretable machine learning

Root concentration factor (RCF) is a significant parameter to characterize uptake and accumulation of hazardous organic contaminants (HOCs) by plant roots. However, complex interactions among chemicals, plant roots and soil make it challenging to identify underlying mechanisms of uptake and accumulation of HOCs. Here, nine machine learning techniques were applied to investigate major factors controlling RCF based on variable combinations of molecular descriptors (MD), MACCS fingerprints, quantum chemistry descriptors (QCD) and three physicochemical properties related to chemical-soil-plant system. Compared to models with variables including MACCS fingerprints or solitary physicochemical properties, the XGBoost-6 model developed by the variable combination of MD, QCD and three physicochemical properties achieved the most remarkable performance, with R2 of 0.977. Model interpretation achieved by permutation variable importance and partial dependence plots revealed the vital importance of HOCs lipophilicity, lipid content of plant roots, soil organic matter content, the overall deformability and the molecular dispersive ability of HOCs for regulating RCF. The integration of MD and QCD with physicochemical properties could improve our knowledge of underlying mechanisms regarding HOCs accumulation in plant roots from innovative structural perspectives. Multiple variables combination-oriented performance improvement of model can be extended to other parameters prediction in environmental risk assessment field.

Apparent quantum yield for photo-production of singlet oxygen in reservoirs and its relation to the water matrix

Man-made reservoirs are important for human daily lives and offer different functions, however they are contaminated due to anthropogenic activities. Dissolved organic matter (DOM) from each reservoir is unique in composition, which further determines its photo-reactivity. Thus, this study aimed to investigate the photo-reactivity of reservoir DOM in terms of the quantum yield for photo-production of singlet oxygen (Ф1O2). We sampled surface water of 50 reservoirs in Japan and determined their Ф1O2 using simulated sunlight together with bulk water analysis. Their Ф1O2 ranged from 1.46 × 10-2 to 6.21 × 10-2 (mean, 2.55 × 10-2), which was identical to those of lakes and rivers reported in the literature, but lower than those of wetland water and wastewater. High-energy triplet-state of DOM accounted for 59.4% of the 1O2 production in the reservoir water on average. Among the bulk water properties, the spectral slope of wavelength from 350 to 400 nm (S350-400) was statistically detected as the most important predictor for Ф1O2. Furthermore, the multiple linear regression model employed S350-400 and the biological index as predictors with no intercorrelations and reasonable accuracy (r2 = 0.86), while the random forest model showed a better accuracy (r2 = 0.90). Overall, these major findings are beneficial for understanding the photo-reactivity of reservoir waters.

Regioselective oxidation of heterocyclic aromatic hydrocarbons catalyzed by cytochrome P450: A case study of carbazole

Recently there are increasing interests in accurately evaluating the health effects of heterocyclic PAHs. However, the activation mechanism and possible metabolites of heterocyclic PAHs catalyzed by human CYP1A1 is still elusive to a great extent. Here, leveraged to high level QM/MM calculations, the corresponding activation pathways of a representative heterocyclic PAHs, carbazole, were systematically explored. The first stage is electrophilic addition or hydrogen abstraction from N-H group. Electrophilic addition was evidenced to be more feasible and regioselectivity at C3 and C4 sites were identified. Correlations between energy barriers and key structural/electrostatic parameters reveal that O-Cα distance and Fe-O-Cα angle are the main origin for the catalytic regioselectivity. Electrophilic addition was determined as the rate-determining step and the subsequent possible reactions include epoxidation, NIH shift (the hydrogen migration from the site of hydroxylation to the adjacent carbon) and proton shuttle. The corresponding products are epoxides, ketones and hydroxylated carbazoles, respectively. The main metabolites (hydroxylated carbazoles) are estimated to be more toxic than carbazole. The regioselectivity of carbazole activated by CYP1A1 is different from the environmental processes (gas and aqueous phase). Collectively, these results will inform the in-depth understanding the metabolic processes of heterocyclic PAHs and aid the accurate evaluation of their health effects.

The application of machine learning to air pollution research: A bibliometric analysis

Machine learning (ML) is an advanced computer algorithm that simulates the human learning process to solve problems. With an explosion of monitoring data and the increasing demand for fast and accurate prediction, ML models have been rapidly developed and applied in air pollution research. In order to explore the status of ML applications in air pollution research, a bibliometric analysis was made based on 2962 articles published from 1990 to 2021. The number of publications increased sharply after 2017, comprising approximately 75% of the total. Institutions in China and United States contributed half of all publications with most research being conducted by individual groups rather than global collaborations. Cluster analysis revealed four main research topics for the application of ML: chemical characterization of pollutants, short-term forecasting, detection improvement and optimizing emission control. The rapid development of ML algorithms has increased the capability to explore the chemical characteristics of multiple pollutants, analyze chemical reactions and their driving factors, and simulate scenarios. Combined with multi-field data, ML models are a powerful tool for analyzing atmospheric chemical processes and evaluating the management of air quality and deserve greater attention in future.

Unified Modeling Approach for Quantifying the Proton and Metal Binding Ability of Soil Dissolved Organic Matter

Soil dissolved organic matter (DOM) is composed of a mass of complex organic compounds in soil solutions and significantly affects a range of (bio)geochemical processes in soil environment. However, how the chemical complexity (i.e., heterogeneity and chemodiversity) of soil DOM molecules affects their proton and metal binding ability remains unclear, which limits our ability for predicting the environmental behavior of DOM and metals. In this study, we developed a unified modeling approach for quantifying the proton and metal binding ability of soil DOM based on Cu titration experiments, Fourier transform ion cyclotron resonance mass spectrometry data, and molecular modeling method. Although soil DOM samples from different regions have enormously heterogeneous and diverse properties, we found that the molecules of soil DOM can be divided into three representative groups according to their Cu binding capacity. Based on the molecular models for individual molecular groups and the relative contributions of each group in each soil DOM, we were able to further develop molecular models for all soil DOM to predict their molecular properties and proton and metal binding ability. Our results will help to develop mechanistic models for predicting the reactivity of soil DOM from various sources.

Machine learning in the identification, prediction and exploration of environmental toxicology: Challenges and perspectives

Over the past few decades, data-driven machine learning (ML) has distinguished itself from hypothesis-driven studies and has recently received much attention in environmental toxicology. However, the use of ML in environmental toxicology remains in the early stages, with knowledge gaps, technical bottlenecks in data quality, high-dimensional/heterogeneous/small-sample data analysis and model interpretability, and a lack of an in-depth understanding of environmental toxicology. Given the above problems, we review the recent progress in the literature and highlight state-of-the-art toxicological studies using ML (such as learning and predicting toxicity in complicated biosystems and multiple-factor environmental scenarios of long-term and large-scale pollution). Beyond predicting simple biological endpoints by integrating untargeted omics and adverse outcome pathways, ML development should focus on revealing toxicological mechanisms. The integration of data-driven ML with other methods (e.g., omics analysis and adverse outcome pathway frameworks) endows ML with widely promising application in revealing toxicological mechanisms. High-quality databases and interpretable algorithms are urgently needed for toxicology and environmental science. Addressing the core issues and future challenges for ML in this review may narrow the knowledge gap between environmental toxicity and computational science and facilitate the control of environmental risk in the future.

Origin of metabolites diversity and selectivity of P450 catalyzed benzo[a]pyrene metabolic activation

Polycyclic Aromatic Hydrocarbon (PAHs) presents one of the most abundant class of environmental pollutants. Recent study shows a lab-synthesized PAHs derivative, helicenium, can selectively kill cancer cells rather than normal cells, calling for the in-depth understanding of the metabolic process. However, the origin of metabolites diversity and selectivity of P450 catalyzed PAHs metabolic activation is still unclear to a great extent. Here we systematically investigated P450 enzymes catalyzed activation mechanism of a representative PAHs, benzo[a]pyrene (BaP), and found the corresponding activation process mainly involves two elementary steps: electrophilic addition and epoxidation. Electrophilic addition step is evidenced to be rate determining step. Two representative binding modes of BaP with P450 were found, which enables the electrophilic addition of Heme (FeO) to almost all the carbons of BaP. This electrophilic addition was proposed to be accelerated by the P450 enzyme environment when compared with the gas phase and water solvent. To dig deeper on the origin of metabolites diversity, we built several linear regression models to explore the structural-energy relationships. The selectivity was eventually attributed to the integrated effects of structural (e.g. O-C distance and O-C-Fe angle) and electrostatic parameters (e.g. charge of C and O) from both BaP and P450.

Endocrine-disrupting metabolic activation of 2-nitrofluorene catalyzed by human cytochrome P450 1A1: A QM/MM approach

Nitropolycyclic aromatic hydrocarbons (NPAHs) present one of the most important airborne pollutants. Recent studies have shown that one of the most abundant NPAHs, 2-Nitrofluorene (NF), was supposed to be converted to endocrine-disrupting metabolites by cytochrome P450 1A1 (CYP1A1) in human cells. However, the mechanism is still largely unexplored. Here the metabolic activation and transformation mechanism of NF catalyzed by CYP1A1 were systematically studied with the aid of Molecular Dynamics, Density Functional Theory and Quantum Mechanics/Molecular Mechanics techniques. We evidence that CYP1A1 can activate NF through two elementary processes: (i) electrophilic addition (12.4 kcal·mol-1) or hydrogen abstraction (38.2 kcal·mol-1) and (ii) epoxidation (5.9 and 8.7 kcal·mol-1) or NIH shift (12.5 and 14.9 kcal·mol-1) or proton shuttle (12.1 kcal·mol-1). Electrophilic addition was found to be the rate-determining step while epoxidation rather than NIH shift or proton shuttle is the more feasible pathway after electrophilic addition. Metabolites 6,7-epoxide-2-nitrofluorene and 7,8-epoxide-2-nitrofluorene were identified as the major epoxidation products. Epoxides are unstable and easy to react with hydrated hydrogen ions and hydroxyls to produce endocrine disrupter 7-hydroxy-2-nitrofluorene. Toxic analysis shows that some of the metabolites are more toxic to model aquatic organisms (e.g. Green algea) than NF. Binding affinity analysis to human sex hormone binding globulin reveals that NF metabolites all have endocrine-disrupting potential. This study provides a comprehensive understanding on the biotransformation process of NF and may aid future studies on various NPAHs activation catalyzed by human P450 enzyme.