Ensemble multiclassification model for predicting developmental toxicity in zebrafish.

In recent years, with the rapid development of society, organic compounds have been released into aquatic environments in various forms, posing a significant threat to the survival of aquatic organisms. The assessment of developmental toxicity is an important part of environmental safety risk systems, helping to identify the potential impacts of organic compounds on the embryonic development of aquatic organisms and enabling early detection and warning of potential ecological risks. Additionally, binary classification models cannot accurately classify organic compounds. Therefore, it is crucial to construct a multiclassification model for predicting the developmental toxicity of organic compounds. In this study, binary and multiclassification models were developed based on the ToxCast™ Phase I chemical library and literature data. The random forest, support vector machine, extreme gradient boosting, adaptive gradient boosting, and C5.0 decision tree algorithms, as well as 8 types of molecular fingerprint were used to establish a multiclassification base model for predicting developmental toxicity through 5-fold cross-validation and external validation. Ultimately, a multiclassification ensemble model was derived through a voting method. The performance of the binary ensemble model, as measured by the balanced accuracy, was 0.918, while that of the multiclassification model was 0.819. The developmental toxicity voting ensemble model (DT-VEM) achieved accuracies of 0.804, 0.834, and 0.855. Furthermore, by utilizing the XGBoost machine learning algorithm to construct separate models for molecular descriptors and substructure molecular fingerprints, we identified several substructures and physical properties related to developmental toxicity. Our research contributes to a more detailed classification of developmental toxicity, providing a new and valuable tool for predicting the developmental toxicity effects of unknown compounds. This supplement addresses the limitations of previous tools, as it offers an enhanced ability to predict potential developmental toxicity in novel compounds.

Effects of dissolved organic matter on the environmental behavior and toxicity of metal nanomaterials: A review.

Metal nanomaterials (MNMs) have been released into the environment during their usage in various products, and their environmental behaviors directly impact their toxicity. Numerous environmental factors potentially affect the behaviors and toxicity of MNMs with dissolved organic matter (DOM) playing the most essential role. Abundant facts showing contradictory results about the effects of DOM on MNMs, herein the occurrence of DOM on the environmental process change of MNMs such as dissolution, dispersion, aggregation, and surface transformation were summarized. We also reviewed the effects of MNMs on organisms and their mechanisms in the environment such as acute toxicity, oxidative stress, oxidative damage, growth inhibition, photosynthesis, reproductive toxicity, and malformation. The presence of DOM had the potential to reduce or enhance the toxicity of MNMs by altering the reactive oxygen species (ROS) generation, dissolution, stability, and electrostatic repulsion of MNMs. Furthermore, we summarized the factors that affected different toxicity including specific organisms, DOM concentration, DOM types, light conditions, detection time, and production methods of MNMs. However, the more detailed mechanism of interaction between DOM and MNMs needs further investigation.

Multi-Endpoint Acute Toxicity Assessment of Organic Compounds Using Large-Scale Machine Learning Modeling.

In recent years, alternative animal testing methods such as computational and machine learning approaches have become increasingly crucial for toxicity testing. However, the complexity and scarcity of available biomedical data challenge the development of predictive models. Combining nonlinear machine learning together with multicondition descriptors offers a solution for using data from various assays to create a robust model. This work applies multicondition descriptors (MCDs) to develop a QSTR (Quantitative Structure-Toxicity Relationship) model based on a large toxicity data set comprising more than 80,000 compounds and 59 different end points (122,572 data points). The prediction capabilities of developed single-task multi-end point machine learning models as well as a novel data analysis approach with the use of Convolutional Neural Networks (CNN) are discussed. The results show that using MCDs significantly improves the model and using them with CNN-1D yields the best result (R2train = 0.93, R2ext = 0.70). Several structural features showed a high level of contribution to the toxicity, including van der Waals surface area (VSA), number of nitrogen-containing fragments (nN+), presence of S-P fragments, ionization potential, and presence of C-N fragments. The developed models can be very useful tools to predict the toxicity of various compounds under different conditions, enabling quick toxicity assessment of new compounds.

Quantitative read-across structure-activity relationship (q-RASAR): A novel approach to estimate the subchronic oral safety (NOAEL) of diverse organic chemicals in rats.

We have developed a quantitative safety prediction model for subchronic repeated doses of diverse organic chemicals on rats using the novel quantitative read-across structure-activity relationship (q-RASAR) approach, which uses similarity-based descriptors for predictive model generation. The experimental -Log (NOAEL) values have been used here as a potential indicator of oral subchronic safety on rats as it determines the maximum dose level for which no observed adverse effects of chemicals are found. A total of 186 data points of diverse organic chemicals have been used for the model generation using structural and physicochemical (0D-2D) descriptors. The read-across-derived similarity, error, and concordance measures (RASAR descriptors) have been extracted from the preliminary 0D-2D descriptors. Then, the combined pool of RASAR and the identified 0D-2D descriptors of the training set were employed to develop the final models by using the partial least squares (PLS) algorithm. The developed PLS model was rigorously validated by various internal and external validation metrics as suggested by the Organization for Economic Co-operation and Development (OECD). The final q-RASAR model is proven to be statistically sound, robust and externally predictive (R2 = 0.85, Q2LOO = 0.82 and Q2F1 = 0.94), superseding the internal as well as external predictivity of the corresponding quantitative structure-activity relationship (QSAR) model as well as previously reported subchronic repeated dose toxicity model found in the literature. In a nutshell, the q-RASAR is an effective approach that has the potential to be used as a good alternative way to improve external predictivity, interpretability, and transferability for subchronic oral safety prediction as well as ecotoxicity risk identification.

Toxicity of individual and mixture of organic compounds to P. Phosphoreum and S. Capricornutum using interpretable simple structural parameters.

Human and environmental ecosystem beings are exposed to multicomponent compound mixtures but the toxicity nature of compound mixtures is not alike to the individual chemicals. This work introduces four models for the prediction of the negative logarithm of median effective concentration (pEC50) of individual chemicals to marine bacteria Photobacterium Phosphoreum (P. Phosphoreum) and algal test species Selenastrum Capricornutum (S. Capricornutum) as well as their mixtures to P. Phosphoreum, and S. Capricornutum. These models provide the simplest approaches for the forecast of pEC50 of some classes of organic compounds from their interpretable structural parameters. Due to the lack of adequate toxicity data for chemical mixtures, the largest available experimental data of individual chemicals (55 data) and their mixtures (99 data) are used to derive the new correlations. The models of individual chemicals are based on two simple structural parameters but chemical mixture models require further interaction terms. The new model's results are compared with the outputs of the best accessible quantitative structure-activity relationships (QSARs) models. Various statistical parameters are done on the new and comparative complex QSAR models, which confirm the higher reliability and simplicity of the new correlations.

In Silico Acute Aquatic Hazard Assessment and Prioritization Using a Grouped Target Site Model: A Case Study of Organic Substances Reported in Permian Basin Hydraulic Fracturing Operations.

Hydraulic fracturing (HF) is commonly used to enhance onshore recovery of oil and gas during production. This process involves the use of a variety of chemicals to support the physical extraction of oil and gas, maintain appropriate conditions downhole (e.g., redox conditions, pH), and limit microbial growth. The diversity of chemicals used in HF presents a significant challenge for risk assessment. The objective of the present study is to establish a transparent, reproducible procedure for estimating 5th percentile acute aquatic hazard concentrations (e.g., acute hazard concentration 5th percentiles [HC5s]) for these substances and validating against existing toxicity data. A simplified, grouped target site model (gTSM) was developed using a database (n = 1696) of diverse compounds with known mode of action (MoA) information. Statistical significance testing was employed to reduce model complexity by combining 11 discrete MoAs into three general hazard groups. The new model was trained and validated using an 80:20 allocation of the experimental database. The gTSM predicts toxicity using a combination of target site water partition coefficients and hazard group-based critical target site concentrations. Model performance was comparable to the original TSM using 40% fewer parameters. Model predictions were judged to be sufficiently reliable and the gTSM was further used to prioritize a subset of reported Permian Basin HF substances for risk evaluation. The gTSM was applied to predict hazard groups, species acute toxicity, and acute HC5s for 186 organic compounds (neutral and ionic). Toxicity predictions and acute HC5 estimates were validated against measured acute toxicity data compiled for HF substances. This case study supports the gTSM as an efficient, cost-effective computational tool for rapid aquatic hazard assessment of diverse organic chemicals. Environ Toxicol Chem 2024;43:1161-1172. © 2024 ExxonMobil Petroleum and Chemical BV. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.

q-RASTR modelling for prediction of diverse toxic chemicals towards T. pyriformis.

A series of diverse organic compounds impose serious detrimental effects on the health of living organisms and the environment. Determination of the structural aspects of compounds that impart toxicity and evaluation of the same is crucial before public usage. The present study aims to determine the structural characteristics of compounds for Tetrahymena pyriformis toxicity using the q-RASTR (Quantitative Read Across Structure-Toxicity Relationship) model. It was developed using RASTR and 2-D descriptors for a dataset of 1792 compounds with defined endpoint (pIGC50) against a model organism, T. pyriformis. For the current study, the whole dataset was divided based on activity/property into the training and test sets, and the q-RASTR model was developed employing six descriptors (three latent variables) having r2, Q2F1 and Q2 values of 0.739, 0.767, and 0.735, respectively. The generated model was thoroughly validated using internationally recognized internal and external validation criteria to assess the model's dependability and predictability. It was highlighted that high molecular weight, aromatic hydroxyls, nitrogen, double bonds, and hydrophobicity increase the toxicity of organic compounds. The current study demonstrates the applicability of the RASTR algorithm in QSTR model development for the prediction of toxic chemicals (pIGC50) towards T. pyriformis.

Unveiling first report on in silico modeling of aquatic toxicity of organic chemicals to Labeo rohita (Rohu) employing QSAR and q-RASAR.

Labeo rohita, a fish species within the Carp family, holds significant dietary and aquacultural importance in South Asian countries. However, the habitats of L. rohita often face exposure to various harmful pesticides and organic compounds originating from industrial and agricultural runoff. It is challenging to individually investigate the effects of each potentially harmful compound. In such cases, in silico techniques like Quantitative Structure-Activity Relationship (QSAR) and quantitative Read-Across Structure-Activity Relationship (q-RASAR) can be employed to construct algorithmic models capable of simultaneously assessing the toxicity of numerous compounds. We utilized the US EPA's ToxValDB database to curate data regarding acute median lethal concentration (LC50) toxicity for L. rohita. The experimental variables included study type (mortality), study duration (ranging from 0.25 h to 4 h), exposure route (static, flowthrough, and renewal), exposure method (drinking water), and types of chemicals (industrial chemicals and pharmaceuticals). Using this dataset, we developed regression-based QSAR and q-RASAR models to predict chemical toxicity to L. rohita based on chemical descriptors. The key descriptors for predicting the toxicity of L. rohita in the regression-based QSAR model include F05[S-Cl], SpMax_EA(ri), s4_relPathLength_2, and SpDiam_AEA(ed). These descriptors can be employed to estimate the toxicity of untested compounds and aid in the development of compounds with lower toxicity based on the presence or absence of these descriptors. Both the QSAR and q-RASAR models serve as valuable tools for understanding the chemicals' structural features responsible for toxicity and for filling gaps in aquatic toxicity data by predicting the toxicity of newly untested compounds in relation to L. rohita. Finally, the developed best model was employed to predict 297 external chemicals, the most toxic substances to L. rohita were identified as cyhalothrin, isobornyl thiocyanatoacetate, and paclobutrzol, while the least toxic ones included ethyl acetate, ethylthiourea, and n-butyric acid.

Read-across-based intelligent learning: development of a global q-RASAR model for the efficient quantitative predictions of skin sensitization potential of diverse organic chemicals.

Environmental chemicals and contaminants cause a wide array of harmful implications to terrestrial and aquatic life which ranges from skin sensitization to acute oral toxicity. The current study aims to assess the quantitative skin sensitization potential of a large set of industrial and environmental chemicals acting through different mechanisms using the novel quantitative Read-Across Structure-Activity Relationship (q-RASAR) approach. Based on the identified important set of structural and physicochemical features, Read-Across-based hyperparameters were optimized using the training set compounds followed by the calculation of similarity and error-based RASAR descriptors. Data fusion, further feature selection, and removal of prediction confidence outliers were performed to generate a partial least squares (PLS) q-RASAR model, followed by the application of various Machine Learning (ML) tools to check the quality of predictions. The PLS model was found to be the best among different models. A simple user-friendly Java-based software tool was developed based on the PLS model, which efficiently predicts the toxicity value(s) of query compound(s) along with their status of Applicability Domain (AD) in terms of leverage values. This model has been developed using structurally diverse compounds and is expected to predict efficiently and quantitatively the skin sensitization potential of environmental chemicals to estimate their occupational and health hazards.

Determining the toxicity of organic compounds to the nematode Caenorhabditis elegans based on aqueous concentrations.

Caenorhabditis elegans is used for assessing the toxicity of chemicals in aqueous medium. However, chemicals can absorb to the bacterial food, which reduces the freely dissolved concentrations of the tested compounds. Thus, based on total or nominal concentrations, toxicity is underestimated, resulting in misleading assumptions on toxicity mechanisms or comparisons to other test organisms. As the verification of freely dissolved exposure concentrations (Cfree) is challenging in small test systems, simple partitioning models might by a good option for estimating Cfree. Therefore, C. elegans was exposed to seven differently acting organic chemicals with varying hydrophobicities, thus also different affinities to bind to the food of C. elegans. Measured concentrations of the dissolved aqueous and the bacterial-bound fraction allowed the calculation of binding constants (Kb). Experimental Kb were comparable to literature data of hydrophobic chemicals and correlated well with their hydrophobicity, expressed as log KOW. The chronic toxicity of the various compounds on C. elegans' reproduction, based on their aqueous concentration, was weakly related to their log KOW. Toxicity expressed based on chemical activity and comparisons with a baseline toxicity model, nevertheless, suggested a narcotic mode of action for most hydrophobic compounds (except methylisothiazolinone and trichlorocarbanilide). Although revealing a similar toxicity ranking than Daphnia magna, C. elegans was less sensitive, probably due to its ability to reduce its internal concentrations by means of its very impermeable cuticle or by efficient detoxification mechanisms. It could be shown that measured aqueous concentrations in the nematode test system corresponded well with freely dissolved concentrations that were modeled using simple mass-balance models from nominal concentrations. This offers the possibility to estimate freely dissolved concentrations of chemicals from nominal concentrations, making routine testing of chemicals and their comparison to other species more accurate.

First report on ecotoxicological QSTR and i-QSTR modeling for the prediction of acute ecotoxicity of diverse organic chemicals against three protozoan species.

The recent years have witnessed an upsurge of interest to assess the toxicity of organic chemicals exhibiting harmful impacts on the environment. In this investigation, we have developed regression-based quantitative structure-toxicity relationship (QSTR) models against three protozoan species (Entosiphon sulcantum, Uronema parduczi, and Chilomonas paramecium) using three sets of descriptor combinations such as ETA indices only, non-ETA descriptors only, and both ETA and non-ETA descriptors to examine the key structural features that determine the toxic properties of protozoa. The interspecies QSTR models (i-QSTRs) were also generated for efficient data gap-filling of toxicity databases. The statistical results of the validated models in terms of both internal and external validation metrics suggested that the models are statistically reliable and robust. Additionally, using these validated models, we screened the DrugBank database containing 11,300 pharmaceuticals for assessing the ecotoxicological properties. The features appearing in the models suggested that non-polar characteristics, electronegativity, hydrogen bonding, π-π, and hydrophobic interactions are responsible for chemical toxicity toward protozoan. The validated models may be utilized for the development of eco-friendly drugs & chemicals, data gap-filling of toxicity databases for regulatory purposes and research, as well as to decrease the use of toxic and hazardous chemicals in the environment.

Application of a developed triple-classification machine learning model for carcinogenic prediction of hazardous organic chemicals to the US, EU, and WHO based on Chinese database.

Cancer, the second largest human disease, has become a major public health problem. The prediction of chemicals' carcinogenicity before their synthesis is crucial. In this paper, seven machine learning algorithms (i.e., Random Forest (RF), Logistic Regression (LR), Support Vector Machines (SVM), Complement Naive Bayes (CNB), K-Nearest Neighbor (KNN), XGBoost, and Multilayer Perceptron (MLP)) were used to construct the carcinogenicity triple classification prediction (TCP) model (i.e., 1A, 1B, Category 2). A total of 1444 descriptors of 118 hazardous organic chemicals were calculated by Discovery Studio 2020, Sybyl X-2.0 and PaDEL-Descriptor software. The constructed carcinogenicity TCP model was evaluated through five model evaluation indicators (i.e., Accuracy, Precision, Recall, F1 Score and AUC). The model evaluation results show that Accuracy, Precision, Recall, F1 Score and AUC evaluation indicators meet requirements (greater than 0.6). The accuracy of RF, LR, XGBoost, and MLP models for predicting carcinogenicity of Category 2 is 91.67%, 79.17%, 100%, and 100%, respectively. In addition, the constructed machine learning model in this study has potential for error correction. Taking XGBoost model as an example, the predicted carcinogenicity level of 1,2,3-Trichloropropane (96-18-4) is Category 2, but the actual carcinogenicity level is 1B. But the difference between Category 2 and 1B is only 0.004, indicating that the XGBoost is one optimum model of the seven constructed machine learning models. Besides, results showed that functional groups like chlorine and benzene ring might influence the prediction of carcinogenic classification. Therefore, considering functional group characteristics of chemicals before constructing the carcinogenicity prediction model of organic chemicals is recommended. The predicted carcinogenicity of the organic chemicals using the optimum machine leaning model (i.e., XGBoost) was also evaluated and verified by the toxicokinetics. The RF and XGBoost TCP models constructed in this paper can be used for carcinogenicity detection before synthesizing new organic substances. It also provides technical support for the subsequent management of organic chemicals.

Support Vector Machine-Based Global Classification Model of the Toxicity of Organic Compounds to Vibrio fischeri.

Vibrio fischeri is widely used as the model species in toxicity and risk assessment. For the first time, a global classification model was proposed in this paper for a two-class problem (Class - 1 with log1/IBC50 ≤ 4.2 and Class + 1 with log1/IBC50 > 4.2, the unit of IBC50: mol/L) by utilizing a large data set of 601 toxicity log1/IBC50 of organic compounds to Vibrio fischeri. Dragon software was used to calculate 4885 molecular descriptors for each compound. Stepwise multiple linear regression (MLR) analysis was used to select the descriptor subset for the models. The ten molecular descriptors used in the classification model reflect the structural information on the Michael-type addition of nucleophiles, molecular branching, molecular size, polarizability, hydrophobic, and so on. Furthermore, these descriptors were interpreted from the point of view of toxicity mechanisms. The optimal support vector machine (SVM) model (C = 253.8 and γ = 0.009) was obtained with the genetic algorithm. The SVM classification model produced a prediction accuracy of 89.1% for the training set (451 log1/IBC50), of 80.0% for the test set (150 log1/IBC50), and of 86.9% for the total data set (601 log1/IBC50), which are higher than that (80.5%, 76%, and 79.4%, respectively) from the binary logistic regression (BLR) model. The global SVM classification model is successful, although it deals with a large data set in relation to the toxicity of organics to Vibrio fischeri.

Data driven toxicity assessment of organic chemicals against Gammarus species using QSAR approach.

Nowadays, organic chemicals play an essential role in almost all walks of life and have become indispensable to modern society. However, the continually synthesized chemicals and the numerous potential adverse endpoints against living organisms increasingly promote the regulators regarding the computational approach as a crucial supplement and an alternative to the traditional animal tests in chemical risk assessment. In this present research, we evaluated the ecotoxicity of chemicals against four typical Gammarus species, which constituted a critical element in detritus cycle and also the recommended species for water monitoring. We first screened the molecular descriptors based on the Genetic Algorithm and then developed the Quantitative Structure-Activity Relationship models using the Multiple Linear Regression method. The statistical results from various validation metrics suggested that the obtained models were internally robust and externally predictive. The application domain analysis based on the leverage approach and standardized residual method demonstrated the broad application range of each model. The interpretation of molecular descriptors in each model suggested that the chemicals with higher polarity and hydrophilicity tend to be less toxic, whereas the lipophilic moieties would enhance the chemical toxicity. Meanwhile, the other selected descriptors, such as Chi-cluster, heterocyclic, and distance matrix descriptors, manifested that the chemical toxicity was also affected by molecular branching, connectivity, electrotopological state, and other various properties. In summary, the present work proposed well-performed QSAR models and clarified the possible toxic mechanism of chemicals against Gammarus species. The obtained models could help predict the toxicity data and conduct a preliminary risk assessment, thus guiding the subsequent animal tests and reducing the assessment cost.

Influence of water exchange rates on toxicity and bioaccumulation of hydrophobic organic chemicals in sediment toxicity tests.

In standardized sediment toxicity tests, the applied water exchange methods range from static to flow-through conditions and vary between protocols and laboratories even for the same test species. This variation potentially results in variable chemical exposure, hampering the interpretation of toxicity and bioaccumulation. To address these issues, we performed sediment toxicity tests with a mixture of three polycyclic aromatic hydrocarbons (PAHs) and the freshwater epibenthic amphipod Hyalella azteca as model chemicals and organism, respectively. Five standardized water exchange methods were applied: static, semi-static, or flow-through conditions. By measuring total (Cdiss) and freely dissolved concentrations (Cfree) of PAHs with water sampling and direct immersion solid-phase microextraction methods, respectively, we found that Cdiss in overlying water differed by a factor of up to 107 among water exchange conditions, whereas both Cdiss and Cfree in pore water did not differ by more than a factor of 2.6. Similar survival rates, growth rates, and bioaccumulation of PAHs between water exchange methods suggest that H. azteca was predominantly exposed to pore water rather than overlying water. By applying mechanistic kinetic modeling to simulate spatiotemporal concentration profiles in sediment toxicity tests, we discuss the importance of the water exchange rates and resulting temporal and spatial exposure variability for the extrapolation of laboratory sediment toxicity to field conditions, particularly for chemicals with relatively low hydrophobicity and sediments with low organic carbon content.

Ensemble multiclassification model for aquatic toxicity of organic compounds.

With environmental pollution becoming increasingly serious, organic compounds have become the main hazard of environmental pollution and exert substantial negative impacts on aquatic organisms. In research pertaining to the acute toxicity of organic compounds, traditional biological experimental methods are time-consuming and expensive. In addition, computer-aided binary classification models cannot accurately classify acute toxicity. Therefore, the multiclassication model is necessary for more accurate classification of acute toxicity. In this study, median lethal concentrations of 373 organic compounds in the environmental toxicology datasets ECOTOX and EAT5 were used. These chemicals were classified into four categories based on the European Economic Community criteria. Then the random forest, support vector machine, extreme gradient boosting, adaptive gradient boosting, and C5.0 decision tree algorithms and eight molecular fingerprints were used to build a multiclassification base model for the acute toxicity of organic compounds. The base models were repeated 100 times with fivefold cross-validation and external validation. The ensemble model was obtained by the voting method. The best base classifier was ExtendFP-C5.0, which had an accuracy, sensitivity and specificity values of 87.30%, 87.32% and 95.76% for external validation, and the voting ensemble model performance of 96.92%, 96.93% and 98.97%, respectively. The ensemble model achieved a higher accuracy than previously reported studies. Our study will help to further classify the acute toxicity of organic compounds to aquatic organisms and predict the hazard classes of organic compounds.

Quasi-SMILES: Self-consistent models for toxicity of organic chemicals to tadpoles.

Simplified molecular input-line entry systems (SMILES) are the representation of the molecular structure that can be used to establish quantitative structure-property/activity relationships (QSPRs/QSARs) for various endpoints expressed as mathematical functions of the molecular architecture. Quasi-SMILES is extending the traditional SMILES by means of additional symbols that reflect experimental conditions. Using the quasi-SMILES models of toxicity to tadpoles gives the possibility to build up models by taking into account the time of exposure. Toxic effects of experimental situations expressed via 188 quasi-SMILES (the negative logarithm of molar concentrations which lead to lethal 50% tadpoles effected during 12 h, 24 h, 48 h, 72 h, and 96 h) were modelled with good results (the average determination coefficient for the validation sets is about 0.97). In this way, we developed new models for this amphibian endpoint, which is poorly studied.

Mixture Risk Assessment of Complex Real-Life Mixtures-The PANORAMIX Project.

Humans are involuntarily exposed to hundreds of chemicals that either contaminate our environment and food or are added intentionally to our daily products. These complex mixtures of chemicals may pose a risk to human health. One of the goals of the European Union's Green Deal and zero-pollution ambition for a toxic-free environment is to tackle the existent gaps in chemical mixture risk assessment by providing scientific grounds that support the implementation of adequate regulatory measures within the EU. We suggest dealing with this challenge by: (1) characterising 'real-life' chemical mixtures and determining to what extent they are transferred from the environment to humans via food and water, and from the mother to the foetus; (2) establishing a high-throughput whole-mixture-based in vitro strategy for screening of real-life complex mixtures of organic chemicals extracted from humans using integrated chemical profiling (suspect screening) together with effect-directed analysis; (3) evaluating which human blood levels of chemical mixtures might be of concern for children's development; and (4) developing a web-based, ready-to-use interface that integrates hazard and exposure data to enable component-based mixture risk estimation. These concepts form the basis of the Green Deal project PANORAMIX, whose ultimate goal is to progress mixture risk assessment of chemicals.

Linking electron ionization mass spectra of organic chemicals to toxicity endpoints through machine learning and experimentation.

Quantitative structure-activity relationship (QSAR) modeling has been widely used to predict the potential harm of chemicals, in which the prediction heavily relies on the accurate annotation of chemical structures. However, it is difficult to determine the accurate structure of an unknown compound in many cases, such as in complex water environments. Here, we solved the above problem by linking electron ionization mass spectra (EI-MS) of organic chemicals to toxicity endpoints through various machine learning methods. The proposed method was verified by predicting 50% growth inhibition of Tetrahymena pyriformis (T. pyriformis) and liver toxicity. The optimal model performance obtained an R2 > 0.7 or balanced accuracy > 0.72 for both the training set and test set. External experimentation further verified the application potential of our proposed method in the toxicity prediction of unknown chemicals. Feature importance analysis allowed us to identify critical spectral features that were responsible for chemical-induced toxicity. Our approach has the potential for toxicity prediction in such fields that it is difficult to determine accurate chemical structures.

Pesticide toxicity towards microalgae increases with environmental mixture complexity.

Effect-directed analysis (EDA) aims at identifying the compound(s) responsible for toxicity in a complex environmental sample where several dozens of contaminants can be present. In this study, we used an environmental mixture extracted from the Polar Organic Chemical Integrative Sampler (POCIS) previously immersed downstream a landfill (River Ponteils, South West France), to perform an EDA approach using a microalgal bioassay based on the photosynthetic capacities of diatom (Nitzschia palea) cultures. Adverse effects on photosynthetic capacities were recorded when algae were exposed to the entire POCIS extract (> 85% inhibition at the highest concentration tested). This result was coherent with the detection of diuron and isoproturon, which were the 2 most concentrated herbicides in the extract. However, the EDA process did not allow pointing out the specific compound(s) responsible for the observed toxicity but rather suggested that multiple compounds were involved in the overall toxicity and caused mixture effects.