Machine Learning-Based Models with High Accuracy and Broad Applicability Domains for Screening PMT/vPvM Substances

Artificial intelligence: A key fulcrum for addressing complex environmental health issues

Environmental health (EH) is a complex and interdisciplinary field dedicated to the examination of environmental behaviours, toxicological effects, health risks, and strategies for mitigating harmful environmental factors. Traditional EH research investigates correlations between risk factors and health outcomes through control variables, but this route is difficult to address complex EH issue. Artificial intelligence (AI) technology not only has accelerated the innovation of the scientific research paradigm but also has become an important tool for solving complex EH problems. However, the in-depth and comprehensive implementation of AI in the field of EH still faces many barriers, such as model generalizability, data privacy protection, algorithm transparency, and regulatory and ethical issues. This review focuses on the compound exposures of EH and explores the potential, challenges, and development directions of AI in four key phases of EH research: (1) data collection, fusion, and management, (2) hazard identification and screening, (3) risk modeling and assessment and (4) EH management. It is not difficult to see that in the future, artificial intelligence technology will inevitably carry out multidimensional simulation of complex exposure factors through multi-mode data fusion, so as to achieve accurate identification of environmental health risks, and eventually become an efficient tool for global environmental health management. This review will help researchers re-examine this strategy and provide a reference for AI to solve complex exposure problems.

Enhanced iodinated disinfection byproducts formation in iodide/iodate-containing water undergoing UV-chloramine sequential disinfection: Machine learning-aided identification of reaction mechanisms

Restricted to the complex nature of dissolved organic matter (DOM) in various aquatic environments, the mechanisms of enhanced iodinated disinfection byproducts (I-DBPs) formation in water containing both I- and IO3- (designated as I-/IO3- in this study) during the ultraviolet (UV)-chloramine sequential disinfection process remains unclear. In this study, four machine learning (ML) models were established to predict I-DBP formation by using DOM and disinfection features as input variables. Extreme gradient boosting (XGB) algorithm outperformed the others in model development using synthetic waters and in cross-dataset generalization of surface waters. Shapley additive explanation (SHAP) analysis, partial dependence plots (PDPs), and individual conditional expectation (ICE) analysis were then employed to explain the models' workings and feature interactions, aiding in identification and quantification of underlying mechanisms. A type of DOM component (namely DC_b) was found as the greatest contributor and identified as reduced quinones associated with broken-down lignin within higher plant-derived fulvic substance, serving as precursors and electron shuttles for I-DBP formation. Based on the interactional effects acquired from explanation results, the ejection of e-aq from excited DOM and pre-existing I- in the I-/IO3- system were identified responsible for the enhanced generation of I-DBPs compared to that in the I- or IO3- alone systems; extra DOM scavenged reactive iodine species (RIS), contributing to a limited enhancement. These findings and the methodology developed here together enhance our understanding of the mechanisms how DOM limitedly promotes I-DBP formation during UV-chloramine sequential disinfection of I-/IO3--containing water and facilitate effective online monitoring in the future.

Entropy Similarity-Driven Transformation Reaction Molecular Networking Reveals Transformation Pathways and Potential Risks of Emerging Contaminants in Wastewater: The Example of Sartans

The transformation pathways and risks of emerging contaminants (ECs) in wastewater remain unclear due to the limited throughput of nontarget screening. In this study, an improved method called entropy similarity-driven transformation reaction molecular networking (ESTRMN) was developed to identify transformation products (TPs) in wastewater. In detail, entropy similarity was the most effective algorithm for identifying parent-product spectrum pairs and a threshold of 0.5 for it was determined with the guarantee of high specificity. Additionally, a TP structure database predicted according to known structures and reactions was established to assist in identification. Sartan is one of the most commonly used angiotensin II receptor blocker antihypertensive drugs. Take sartans as an example, 69 TPs of sartans with confidence levels above 3 were identified by ESTRMN, 43 of which were newly discovered. The most common reactions included hydroxylation, hydrolysis, and oxidation, resulting in the majority of sartan TPs exhibiting higher persistence, mobility, and toxicity (PMT) than their parents. The concentration of 75% sartans and TPs increased after treatment in a WWTP, and the overall risk has not been effectively mitigated. This study emphasizes the role of ESTRMN in incorporating TPs of ECs into environmental monitoring protocols and risk assessment frameworks for wastewater management.

Predictive modelling of peroxisome proliferator-activated receptor gamma (PPARγ) IC50 inhibition by emerging pollutants using light gradient boosting machine

Peroxisome proliferator-activated receptor gamma (PPARγ), a critical nuclear receptor, plays a pivotal role in regulating metabolic and inflammatory processes. However, various environmental contaminants can disrupt PPARγ function, leading to adverse health effects. This study introduces a novel approach to predict the inhibitory activity (IC50 values) of 140 chemical compounds across 13 categories, including pesticides, organochlorines, dioxins, detergents, flame retardants, and preservatives, on PPARγ. The predictive model, based on the light-gradient boosting machine (LightGBM) algorithm, was trained on a dataset of 1804 molecules showed r2 values of 0.82 and 0.59, Mean Absolute Error (MAE) of 0.38 and 0.58, and Root Mean Square Error (RMSE) of 0.54 and 0.76 for the training and test sets, respectively. This study provides novel insights into the interactions between emerging contaminants and PPARγ, highlighting the potential hazards and risks these chemicals may pose to public health and the environment. The ability to predict PPARγ inhibition by these hazardous contaminants demonstrates the value of this approach in guiding enhanced environmental toxicology research and risk assessment.

Construction of interpretable ensemble learning models for predicting bioaccumulation parameters of organic chemicals in fish

Accurate prediction of bioaccumulation parameters is essential for assessing exposure, hazards, and risks of chemicals. However, the majority of prediction models on bioaccumulation parameters are individual models based on a single algorithm and lack model interpretation, resulting in unsatisfactory prediction accuracy due to inherent constraints of the algorithm and weak interpretability. Ensemble learning (EL) that combine multiple algorithms, coupled with SHapley Additive exPlanation (SHAP) method, may overcome the limitations. Herein, EL models were constructed for three bioaccumulation parameters using datasets covering 2496 chemicals. The EL models demonstrated superior prediction accuracy compared to both individual models developed in this study and those from previous research, achieving a coefficient of determination of up to 0.861 on the validation sets. Applicability domains were characterized using a structure-activity landscape-based (abbreviated as ADSAL) methodology. The optimal EL models, together with the ADSAL, were successfully used to predict bioaccumulation parameters for 4374 chemicals included in the Inventory of Existing Chemical Substances of China. Model interpretation using the SHAP method offered insight into key features influencing bioaccumulation potential, including hydrophobicity, water solubility, polarizability, ionization potential, weight, and volume of molecules. Overall, the study provides data and models to support the sound management and risk assessment of chemicals.

Transfer Learning with a Graph Attention Network and Weighted Loss Function for Screening of Persistent, Bioaccumulative, Mobile, and Toxic Chemicals

In silico methods for screening hazardous chemicals are necessary for sound management. Persistent, bioaccumulative, mobile, and toxic (PBMT) chemicals persist in the environment and have high mobility in aquatic environments, posing risks to human and ecological health. However, lack of experimental data for the vast number of chemicals hinders identification of PBMT chemicals. Through an extensive search of measured chemical mobility data, as well as persistent, bioaccumulative, and toxic (PBT) chemical inventories, this study constructed comprehensive data sets on PBMT chemicals. To address the limited volume of the PBMT chemical data set, a transfer learning (TL) framework based on graph attention network (GAT) architecture was developed to construct models for screening PBMT chemicals, designating the PBT chemical inventories as source domains and the PBMT chemical data set as target domains. A weighted loss (LW) function was proposed and proved to mitigate the negative impact of imbalanced data on the model performance. Results indicate the TL-GAT models outperformed GAT models, along with large coverage of applicability domains and interpretability. The constructed models were employed to identify PBMT chemicals from inventories consisting of about 1 × 106 chemicals. The developed TL-GAT framework with the LW function holds broad applicability across diverse tasks, especially those involving small and imbalanced data sets.

Determining high priority disinfection byproducts based on experimental aquatic toxicity data and predictive models: Virtual screening and in vivo study

Only about 100 disinfection byproducts (DBPs) have been tested for their potential aquatic toxicity. It is not known which specific DBPs, DBP main groups, and DBP subgroups are more toxic due to the lack of experimental toxicity data. Herein, high priority specific DBPs, DBP main groups, DBP subgroups, most sensitive model aquatic species, potential PBT and PMT (persistent, bioaccumulative/mobile, and toxic) DBPs were virtually screened for 1187 updated DBPs inventory. Priority setting based on experimental and predicted acute and chronic aquatic toxicity data found that the aromatic and alicyclic DBPs in four DBPs main groups showed high priority because larger proportions of aromatic and alicyclic DBPs are in high hazard categories (i.e. Acute and/or Chronic Toxic-1 or Toxic-2) according to the criteria in GHS system compared to the aliphatic and heterocyclic DBPs. The halophenols, estrogen-DBPs, nonhalogenated esters, and nonhalogenated aldehydes were recognized as high priority DBPs subgroups. For specific DBPs, 19 and 31 DBPs should be highly concerned in the future study because both acute and chronic toxicity of those DBPs to all of the three aquatic life (algae, Daphnia magna, fish) were classified as Toxic-1 and Toxic-2, respectively. The Daphnia magna and algae were sensitive to the acute toxicity of DBPs, while the fish and Daphnia magna were sensitive to the chronic toxicity of DBPs. One potential PBT (Tetrachlorobisphenol A) and four potential PMT DBPs were identified. For verification, the acute toxicity of four DBPs on three aquatic organism were performed, and their tested acute toxicity data to three aquatic organisms were consistent with the predictions. Our results could be beneficial to government regulators to adopt effective measures to limit the discharge of high priority DBPs and help the scientific community to develop or improve disinfection processes to reduce the production of high priority DBPs.

Integrated Transfer Learning and Multitask Learning Strategies to Construct Graph Neural Network Models for Predicting Bioaccumulation Parameters of Chemicals

Accurate prediction of parameters related to the environmental exposure of chemicals is crucial for the sound management of chemicals. However, the lack of large data sets for training models may result in poor prediction accuracy and robustness. Herein, integrated transfer learning (TL) and multitask learning (MTL) was proposed for constructing a graph neural network (GNN) model (abbreviated as TL-MTL-GNN model) using n-octanol/water partition coefficients as a source domain. The TL-MTL-GNN model was trained to predict three bioaccumulation parameters based on enlarged data sets that cover 2496 compounds with at least one bioaccumulation parameter. Results show that the TL-MTL-GNN model outperformed single-task GNN models with and without the TL, as well as conventional machine learning models trained with molecular descriptors or fingerprints. Applicability domains were characterized by a state-of-the-art structure-activity landscape-based (abbreviated as ADSAL) methodology. The TL-MTL-GNN model coupled with the optimal ADSAL was employed to predict bioaccumulation parameters for around 60,000 chemicals, with more than 13,000 compounds identified as bioaccumulative chemicals. The high predictive accuracy and robustness of the TL-MTL-GNN model demonstrate the feasibility of integrating the TL and MTL strategy in modeling small-sized data sets. The strategy holds significant potential for addressing small data challenges in modeling environmental chemicals.

Development and application of machine learning models for prediction of soil available cadmium based on soil properties and climate features

Identifying the key influencing factors in soil available cadmium (Cd) is crucial for preventing the Cd accumulation in the food chain. However, current experimental methods and traditional prediction models for assessing available Cd are time-consuming and ineffective. In this study, machine learning (ML) models were developed to investigate the intricate interactions among soil properties, climate features, and available Cd, aiming to identify the key influencing factors. The optimal model was obtained through a combination of stratified sampling, Bayesian optimization, and 10-fold cross-validation. It was further explained through the utilization of permutation feature importance, 2D partial dependence plot, and 3D interaction plot. The findings revealed that pH, surface pressure, sensible heat net flux and organic matter content significantly influenced the Cd accumulation in the soil. By utilizing historical soil surveys and climate change data from China, this study predicted the spatial distribution trend of available Cd in the Chinese region, highlighting the primary areas with heightened Cd activity. These areas were primarily located in the eastern, southern, central, and northeastern China. This study introduces a novel methodology for comprehending the process of available Cd accumulation in soil. Furthermore, it provides recommendations and directions for the remediation and control of soil Cd pollution.

Occurrence and Ecological Risk of Alkylamine Triazines in Chinese Estuarine Sediments: An Emerging Class of Persistent, Mobile, and Toxic Substances

Identifying persistent, mobile, and toxic (PMT) substances from synthetic chemicals is critical for chemical management and ecological risk assessment. Inspired by the triazine analogues (e.g., atrazine and melamine) in the original European Union's list of PMT substances, the occurrence and compositions of alkylamine triazines (AATs) in the estuarine sediments of main rivers along the eastern coast of China were comprehensively explored by an integrated strategy of target, suspect, and nontarget screening analysis. A total of 44 AATs were identified, of which 23 were confirmed by comparison with authentic standards. Among the remaining tentatively identified analogues, 18 were emerging pollutants not previously reported in the environment. Tri- and di-AATs were the dominant analogues, and varied geographic distributions of AATs were apparent in the investigated regions. Toxic unit calculations indicated that there were acute and chronic risks to algae from AATs on a large geographical scale, with the antifouling biocide cybutryne as a key driver. The assessment of physicochemical properties further revealed that more than half of the AATs could be categorized as potential PMT and very persistent and very mobile substances at the screening level. These results highlight that AATs are a class of PMT substances posing high ecological impacts on the aquatic environment and therefore require more attention.

Graph Convolutional Network-Enhanced Model for Screening Persistent, Mobile, and Toxic and Very Persistent and Very Mobile Substances

The global management for persistent, mobile, and toxic (PMT) and very persistent and very mobile (vPvM) substances has been further strengthened with the rapid increase of emerging contaminants. The development of a ready-to-use and publicly available tool for the high-throughput screening of PMT/vPvM substances is thus urgently needed. However, the current model building with the coupling of conventional algorithms, small-scale data set, and simplistic features hinders the development of a robust model for screening PMT/vPvM with wide application domains. Here, we construct a graph convolutional network (GCN)-enhanced model with feature fusion of a molecular graph and molecular descriptors to effectively utilize the significant correlation between critical descriptors and PMT/vPvM substances. The model is built with 213,084 substances following the latest PMT classification criteria. The application domains of the GCN-enhanced model assessed by kernel density estimation demonstrate the high suitability for high-throughput screening PMT/vPvM substances with both a high accuracy rate (86.6%) and a low false-negative rate (6.8%). An online server named PMT/vPvM profiler is further developed with a user-friendly web interface (http://www.pmt.zj.cn/). Our study facilitates a more efficient evaluation of PMT/vPvM substances with a globally accessible screening platform.

Prediction of molecular-specific mutagenic alerts and related mechanisms of chemicals by a convolutional neural network (CNN) model based on SMILES split

Structural alerts (SAs) are essential to identify chemicals for toxicity evaluation and health risk assessment. We constructed a novel SMILES split-based deep learning model (SSDL) that was trained and verified with 5850 chemicals from the ISSSTY database and 384 external test chemicals from published papers. The training accuracy was above 0.90 and the evaluation metrics (precision, recall and F1-score) all reached 0.78 or above on both internal and external test chemicals. In this model, the molecular-specific fragment importance of chemicals was first quantified independently. Then, the SA identification method based on the importance of these fragments was statistically analyzed and verified with the ISSSTY test and external test chemicals containing one of 28 typical SAs, and most of the performances were better than that of expert rules. Furthermore, a mutagenicity mechanism prediction method was developed using 237 chemicals with four known mutagenic mechanisms based on molecular similarity calibrated by the SSDL method and fragment importance, which significantly improved accuracy in three mechanisms and had comparable accuracy in the other one compared to traditional methods. Overall, the SSDL model quantifying fragment toxicity within molecules would be a novel potentially powerful tool in the determination and visualization of molecular-specific SAs and the prediction of mutagenicity mechanisms for environmental or industrial compounds and drugs.

Machine learning coupled with causal inference to identify COVID-19 related chemicals that pose a high concern to drinking water

Various synthetic substances were utilized in large quantities during the recent coronavirus pandemic, COVID-19. Some of these chemicals could potentially enter drinking water sources. Persistent, mobile, and toxic (PMT) substances have been recognized as a threat to drinking water resources. It has not yet been assessed how many COVID-19 related substances could be considered PMT substances. One reason is the lack of high-quality experimental data for the identification of PMT substances. To solve this problem, we applied a machine learning model to identify the PMT substances among COVID-19 related chemicals. The optimal model achieved an accuracy of 90.6% based on external test data. The model interpretation and causal inference indicated that our approach understood causation between PMT properties and molecular descriptors. Notably, the screening results showed that over 60% of the COVID-19 chemicals considered are candidate PMT substances, which should be prioritized to prevent undue pollution of water resources.

MatGPT: A Vane of Materials Informatics from Past, Present, to Future

Combining materials science, artificial intelligence (AI), physical chemistry, and other disciplines, materials informatics is continuously accelerating the vigorous development of new materials. The emergence of "GPT (Generative Pre-trained Transformer) AI" shows that the scientific research field has entered the era of intelligent civilization with "data" as the basic factor and "algorithm + computing power" as the core productivity. The continuous innovation of AI will impact the cognitive laws and scientific methods, and reconstruct the knowledge and wisdom system. This leads to think more about materials informatics. Here, a comprehensive discussion of AI models and materials infrastructures is provided, and the advances in the discovery and design of new materials are reviewed. With the rise of new research paradigms triggered by "AI for Science", the vane of materials informatics: "MatGPT", is proposed and the technical path planning from the aspects of data, descriptors, generative models, pretraining models, directed design models, collaborative training, experimental robots, as well as the efforts and preparations needed to develop a new generation of materials informatics, is carried out. Finally, the challenges and constraints faced by materials informatics are discussed, in order to achieve a more digital, intelligent, and automated construction of materials informatics with the joint efforts of more interdisciplinary scientists.

End-to-end machine-learning for high-gravity ammonia stripping: Bridging the gap between scientific research and user-friendly applications

The removal and recovery of ammonia from wastewater are critical processes for achieving global environmental sustainability and promoting circular economic development. High-gravity technology is an advanced solution to achieve ammonia stripping from wastewater. This study used machine-learning (ML) techniques to provide more comprehensive insights on various influencing factors, including the operating parameters, wastewater characteristics, and design parameters of rotating packed beds. Bayesian auto-optimization combined with a boosting algorithm effectively overcame the challenges of modeling complex datasets with small sample sizes, multidimensional data, missing values, and skewed distributions. Accurate ML based predictive models for the ammonia removal efficiency (η) and mass transfer coefficient (KLa) were developed, the performance on the training set was R2 = 0.98 and R2 = 0.89, and on the testing set was R2 = 0.98 and R2 = 0.82. The developed model revealed that the stripping stage and gas-liquid ratio were the most influential features for predicting η, whereas the liquid flow and high-gravity factor were the most important features for predicting KLa. The well-trained model was then deployed in an online software application that could provide both predictive and auto-update functions for operators and managers, ensuring that practitioners could use the model. The end-to-end machine-learning approach used in this study-that is, covering data collection, model development, and application-could improve the availability of research results, providing valuable references for the further advancement of technology in the field of environmental.

PBT assessment of chemicals detected in effluent of wastewater treatment plants by suspected screening analysis

Wastewater treatment plants (WWTPs) are the major sources of contaminants discharged into downstream water bodies. Profiling the contaminants in effluent of WWTPs is crucial to assess the potential eco-risks toward downstream organisms. To this end, this study investigated the contaminants in effluent of 10 WWTPs locating in 10 cities of Yangtze River delta region of China by suspected screening analysis. Further, the persistence, bioaccumulation, toxicity (PBT) and the characteristics sub-structures of PBT-like chemicals were analyzed. Totally, 704 chemicals including 155 chemical products, 31 food additives, 52 natural substances, 112 personal care products, 123 pesticides, 192 pharmaceuticals, 17 hormones and 22 others were found. The results of PBT analysis suggested that 42 chemicals (5.97% among the detected chemicals in WWTPs) were with PBT property. Among them, 31 contaminants were not reported previously. 9 characteristics sub-structures (N-methyleneisobutylamine, 1-naphthaldehyde, 2,3,3-trimethylcyclohexene, cyclohexanol, N-sec-butyl-n-propylamine, (5E)-2,6-dimethylocta-1,5-diene, 2-ethylphenol, pentadecane and 6-methoxyhexane) were found for PBT-like chemicals. The sub-structures of highly linear alkyl partially explained the significantly higher PBT score for personal care products. Present study provides fundamental information on PBT properties of contaminants in effluent of WWTPs, which will benefit to prioritize contaminants with high concerns in effluent of WWTPs.

Applicability Domains Based on Molecular Graph Contrastive Learning Enable Graph Attention Network Models to Accurately Predict 15 Environmental End Points

In silico models for predicting physicochemical properties and environmental fate parameters are necessary for the sound management of chemicals. This study employed graph attention network (GAT) algorithms to construct such models on 15 end points. The results showed that the GAT models outperformed the previous state-of-the-art models, and their performance was not influenced by the presence or absence of compounds with certain structures. Molecular similarity density (ρs) was found to be a key metrics characterizing data set modelability, in addition to the proportion of compounds at activity cliffs. By introducing molecular graph (MG) contrastive learning, MG-based ρs and molecular inconsistency in activities (IA) were calculated and employed for characterizing the structure-activity landscape (SAL)-based applicability domain ADSAL{ρs, IA}. The GAT models coupled with ADSAL{ρs, IA} significantly improved the prediction coefficient of determination (R2) on all the end points by an average of 14.4% and enabled all the end points to have R2 > 0.9, which could hardly be achieved previously. The models were employed to screen persistent, mobile, and/or bioaccumulative chemicals from inventories consisting of about 106 chemicals. Given the current state-of-the-art model performance and coverage of the various environmental end points, the constructed models with ADSAL{ρs, IA} may serve as benchmarks for future efforts to improve modeling efficacy.

Advances and applications of machine learning and deep learning in environmental ecology and health

Machine learning (ML) and deep learning (DL) possess excellent advantages in data analysis (e.g., feature extraction, clustering, classification, regression, image recognition and prediction) and risk assessment and management in environmental ecology and health (EEH). Considering the rapid growth and increasing complexity of data in EEH, it is of significance to summarize recent advances and applications of ML and DL in EEH. This review summarized the basic processes and fundamental algorithms of the ML and DL modeling, and indicated the urgent needs of ML and DL in EEH. Recent research hotspots such as environmental ecology and restoration, environmental fate of new pollutants, chemical exposures and risks, chemical hazard identification and control were highlighted. Various applications of ML and DL in EEH demonstrate their versatility and technological revolution, and present some challenges. The perspective of ML and DL in EEH were further outlined to promote the innovative analysis and cultivation of the ML-driven research paradigm.

Dynamic Source Distribution and Emission Inventory of a Persistent, Mobile, and Toxic (PMT) Substance, Melamine, in China

Persistent, mobile, and toxic (PMT) substances are affecting the safety of drinking water and are threatening the environment and human health. Many PMT substances are used in industrial processing or consumer products, but their sources and emissions mostly remain unclear. This study presents a long-term source distribution and emission estimation of melamine, a high-production-volume PMT substance of emerging global concern. The results indicate that in China, approximately 1858.7 kilotonnes (kt) of melamine were released into the water (∼58.9%), air (∼27.0%), and soil systems (∼14.1%) between 1995 and 2020, mainly from its production and use in the decorative panels, textiles, and paper industries. The textile and paper industries have the highest emission-to-consumption ratios, with more than 90% emissions per unit consumption. Sewage treatment plants are the largest source of melamine in the environment for the time being, but in-use products and their wastes will serve as significant melamine sources in the future. The study prompts priority action to control the risk of PMT substances internationally.

Developing machine learning approaches to identify candidate persistent, mobile and toxic (PMT) and very persistent and very mobile (vPvM) substances based on molecular structure

Determining which substances on the global market could be classified as persistent, mobile and toxic (PMT) substances or very persistent, very mobile (vPvM) substances is essential to prevent or reduce drinking water contamination from them. This study developed machine learning models based on different molecular descriptors (MDs) and defined applicability domains for the screening of PMT/vPvM substances. The models were trained with 3111 substances with expert weight-of-evidence based PMT/vPvM hazard classifications that considered the highest quality data available. The model was based on the hypothesis that PMT/vPvM substances contain similar MDs, representative of chemical structures resistant to degradation, be associated with low sorption (or high-water solubility) and in some cases be associated with known toxic mechanisms. All possible model combinations were tested by integrating different molecular description methods, data balancing strategies and machine learning algorithms. Our model allows one-step prediction of candidate PMT/vPvM substances, and our method was compared with the approach predicting P, M and T separately (i.e. three-step prediction). The results showed that the one-step model achieved a higher accuracy of 92% for PMT/vPvM identification (i.e. positive samples) for an internal test set, and also resulted in a higher accuracy of 90% for an external test set of chemical pollutants detected in Taihu Lake, China. Furthermore, prediction mechanism of the model was interpreted by Shapley additive explanations (SHAP). This work presents an advance of big data in silico screening models for the identification of substances that potentially meet the PMT/vPvM criteria.

Physics-informed neural network-based serial hybrid model capturing the hidden kinetics for sulfur-driven autotrophic denitrification process

Sulfur-driven autotrophic denitrification (SdAD) is a biological process that can remove nitrate from low carbon/nitrogen (C/N) ratio wastewater. Although this process has been intensively researched, the mechanism whereby its intermediates (i.e., elemental sulfur and nitrite ions) are generated and accumulated remains elusive. Existing mathematical models developed for SdAD cannot accurately predict the intermediates in SdAD because of the incomplete knowledge of process kinetic resulting from changes in the environmental conditions and electron competition during SdAD. To address this limitation, we proposed a novel serial hybrid model structure based on a physics-informed neural network (PINN) to capture the dynamics of the process kinetics and predict the substrate concentrations in SdAD. In this study, we evaluated the model through numerical experiments and applied it to real case studies involving batch and continuous-flow reactor scenarios. By leveraging the PINN approach, the hybrid model yielded accurate predictions at both the state (i.e. substrate concentration) and kinetic levels in the numerical experiments and performed better than both mechanistic and purely data-driven models in the case studies. Furthermore, we used the trained hybrid model to design control strategies for SdAD and a novel integrated process involving SdAD and anammox for energy-efficient nitrogen removal. Finally, we discuss the advantages and application scope of the PINN-based hybrid model.

Machine Learning Model for Screening Thyroid Stimulating Hormone Receptor Agonists Based on Updated Datasets and Improved Applicability Domain Metrics

Machine learning (ML) models for screening endocrine-disrupting chemicals (EDCs), such as thyroid stimulating hormone receptor (TSHR) agonists, are essential for sound management of chemicals. Previous models for screening TSHR agonists were built on imbalanced datasets and lacked applicability domain (AD) characterization essential for regulatory application. Herein, an updated TSHR agonist dataset was built, for which the ratio of active to inactive compounds greatly increased to 1:2.6, and chemical spaces of structure-activity landscapes (SALs) were enhanced. Resulting models based on 7 molecular representations and 4 ML algorithms were proven to outperform previous ones. Weighted similarity density (ρ_s) and weighted inconsistency of activities (I_A) were proposed to characterize the SALs, and a state-of-the-art AD characterization methodology AD_SAL{ρ_s, I_A} was established. An optimal classifier developed with PubChem fingerprints and the random forest algorithm, coupled with AD_SAL{ρ_s ≥ 0.15, I_A ≤ 0.65}, exhibited good performance on the validation set with the area under the receiver operating characteristic curve being 0.984 and balanced accuracy being 0.941 and identified 90 TSHR agonist classes that could not be found previously. The classifier together with the AD_SAL{ρ_s, I_A} may serve as efficient tools for screening EDCs, and the AD characterization methodology may be applied to other ML models.

Characterization of oxidative damage induced by nanoparticles via mechanism-driven machine learning approaches

The wide application of TiO₂-based engineered nanoparticles (nTiO₂) inevitably led to release into aquatic ecosystems. Importantly, increasing studies have emphasized the high risks of nTiO₂ to coastal environments. Bivalves, the representative benthic filter feeders in coastal zones, acted as important roles to assess and monitor the toxic effects of nanoparticles. Oxidative damage was one of the main toxic mechanisms of nTiO₂ on bivalves, but the experimental variables/nanomaterial characteristics were diverse and the toxicity mechanism was complex. Therefore, it was very necessary to develop machine learning model to characterize and predict the potential toxicity. In this study, thirty-six machine learning models were built by nanodescriptors combined with six machine learning algorithms. Among them, random forest (RF) - catalase (CAT), k-neighbors classifier (KNN) - glutathione peroxidase (GPx), neural networks - multilayer perceptron (ANN) - glutathione s-transferase (GST), random forest (RF) - malondialdehyde (MDA), random forest (RF) - reactive oxygen species (ROS), and extreme gradient boosting decision tree (XGB) - superoxide dismutase (SOD) models performed good with high accuracy and balanced accuracy for both training sets and external validation sets. Furthermore, the best model revealed the predominant factors (exposure concentration, exposure periods, and exposure matrix) influencing the oxidative stress induced by nTiO₂. These results showed that high exposure concentrations and short exposure-intervals tended to cause oxidative damage to bivalves. In addition, gills and digestive glands could be vulnerable to nTiO₂-induced oxidative damage as tissues/organs differences were the important factors controlling MDA activity. This study provided insights into important nano-features responsible for the different indicators of oxidative stress and thereby extended the application of machine learning approaches in toxicological assessment for nanoparticles.