Uncertainty-Informed Deep Transfer Learning of Perfluoroalkyl and Polyfluoroalkyl Substance Toxicity

Expanding Predictive Capacities in Toxicology: Insights from Hackathon-Enhanced Data and Model Aggregation

In the realm of predictive toxicology for small molecules, the applicability domain of QSAR models is often limited by the coverage of the chemical space in the training set. Consequently, classical models fail to provide reliable predictions for wide classes of molecules. However, the emergence of innovative data collection methods such as intensive hackathons have promise to quickly expand the available chemical space for model construction. Combined with algorithmic refinement methods, these tools can address the challenges of toxicity prediction, enhancing both the robustness and applicability of the corresponding models. This study aimed to investigate the roles of gradient boosting and strategic data aggregation in enhancing the predictivity ability of models for the toxicity of small organic molecules. We focused on evaluating the impact of incorporating fragment features and expanding the chemical space, facilitated by a comprehensive dataset procured in an open hackathon. We used gradient boosting techniques, accounting for critical features such as the structural fragments or functional groups often associated with manifestations of toxicity.

An artificial intelligence platform for automated PFAS subgroup classification: A discovery tool for PFAS screening

Since structural analyses and toxicity assessments have not been able to keep up with the discovery of unknown per- and polyfluoroalkyl substances (PFAS), there is an urgent need for effective categorization and grouping of PFAS. In this study, we presented PFAS-Atlas, an artificial intelligence-based platform containing a rule-based automatic classification system and a machine learning-based grouping model. Compared with previously developed classification software, the platform's classification system follows the latest Organization for Economic Co-operation and Development (OECD) definition of PFAS and reduces the number of uncategorized PFAS. In addition, the platform incorporates deep unsupervised learning models to visualize the chemical space of PFAS by clustering similar structures and linking related classes. Through real-world use cases, we demonstrate that PFAS-Atlas can rapidly screen for relationships between chemical structure and persistence, bioaccumulation, or toxicity data for PFAS. The platform can also guide the planning of the PFAS testing strategy by showing which PFAS classes urgently require further attention. Ultimately, the release of PFAS-Atlas will benefit both the PFAS research and regulation communities.

Elucidating Key Characteristics of PFAS Binding to Human Peroxisome Proliferator-Activated Receptor Alpha: An Explainable Machine Learning Approach

Per- and polyfluoroalkyl substances (PFAS) are widely employed anthropogenic fluorinated chemicals known to disrupt hepatic lipid metabolism by binding to human peroxisome proliferator-activated receptor alpha (PPARα). Therefore, screening for PFAS that bind to PPARα is of critical importance. Machine learning approaches are promising techniques for rapid screening of PFAS. However, traditional machine learning approaches lack interpretability, posing challenges in investigating the relationship between molecular descriptors and PPARα binding. In this study, we aimed to develop a novel, explainable machine learning approach to rapidly screen for PFAS that bind to PPARα. We calculated the PPARα-PFAS binding score and 206 molecular descriptors for PFAS. Through systematic and objective selection of important molecular descriptors, we developed a machine learning model with good predictive performance using only three descriptors. The molecular size (b_single) and electrostatic properties (BCUT_PEOE_3 and PEOE_VSA_PPOS) are important for PPARα-PFAS binding. Alternative PFAS are considered safer than their legacy predecessors. However, we found that alternative PFAS with many carbon atoms and ether groups exhibited a higher affinity for PPARα. Therefore, confirming the toxicity of these alternative PFAS compounds with such characteristics through biological experiments is important.

Research Priorities for the Environmental Risk Assessment of Per- and Polyfluorinated Substances

Per- and polyfluorinated substances (PFAS) are a group of thousands of ubiquitously applied persistent industrial chemicals. The field of PFAS environmental research is developing rapidly, but suffers from substantial biases toward specific compounds, environmental compartments, and organisms. The aim of our study was therefore to highlight current developments and to identify knowledge gaps and subsequent research needs that would contribute to a comprehensive environmental risk assessment for PFAS. To this end, we consulted the open literature and databases and found that knowledge of the environmental fate of PFAS is based on the analysis of <1% of the compounds categorized as PFAS. Moreover, soils and suspended particulate matter remain largely understudied. The bioavailability, bioaccumulation, and food web transfer studies of PFAS also focus on a very limited number of compounds and are biased toward aquatic biota, predominantly fish, and less frequently aquatic invertebrates and macrophytes. The available ecotoxicity data revealed that only a few PFAS have been well studied for their environmental hazards, and that PFAS ecotoxicity data are also strongly biased toward aquatic organisms. Ecotoxicity studies in the terrestrial environment are needed, as well as chronic, multigenerational, and community ecotoxicity research, in light of the persistency and bioaccumulation of PFAS. Finally, we identified an urgent need to unravel the relationships among sorption, bioaccumulation, and ecotoxicity on the one hand and molecular descriptors of PFAS chemical structures and physicochemical properties on the other, to allow predictions of exposure, bioaccumulation, and toxicity. Environ Toxicol Chem 2023;42:2302-2316. © 2023 The Authors. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.

A Review on Per- and Polyfluoroalkyl Substances in Pregnant Women: Maternal Exposure, Placental Transfer, and Relevant Model Simulation

Per- and polyfluoroalkyl substances (PFASs) are important and ubiquitous environmental contaminants worldwide. These novel contaminants can enter human bodies via various pathways, subsequently posing risks to the ecosystem and human health. The exposure of pregnant women to PFASs might pose risks to the health of mothers and the growth and development of fetuses. However, little information is available about the placental transfer of PFASs from mothers to fetuses and the related mechanisms through model simulation. In the present study, based upon a review of previously published literature, we initially summarized the exposure pathways of PFASs in pregnant women, factors affecting the efficiency of placental transfer, and mechanisms associated with placental transfer; outlined simulation analysis approaches using molecular docking and machine learning to reveal the mechanisms of placental transfer; and finally highlighted future research emphases that need to be focused on. Consequently, it was notable that the binding of PFASs to proteins during placental transfer could be simulated by molecular docking and that the placental transfer efficiency of PFASs could also be predicted by machine learning. Therefore, future research on the maternal-fetal transfer mechanisms of PFASs with the benefit of simulation analysis approaches is warranted to provide a scientific basis for the health effects of PFASs on newborns.

Structural Analysis and Prediction of Hematotoxicity Using Deep Learning Approaches

Hematotoxicity has been becoming a serious but overlooked toxicity in drug discovery. However, only a few in silico models have been reported for the prediction of hematotoxicity. In this study, we constructed a high-quality dataset comprising 759 hematotoxic compounds and 1623 nonhematotoxic compounds and then established a series of classification models based on a combination of seven machine learning (ML) algorithms and nine molecular representations. The results based on two data partitioning strategies and applicability domain (AD) analysis illustrate that the best prediction model based on Attentive FP yielded a balanced accuracy (BA) of 72.6%, an area under the receiver operating characteristic curve (AUC) value of 76.8% for the validation set, and a BA of 69.2%, an AUC of 75.9% for the test set. In addition, compared with existing filtering rules and models, our model achieved the highest BA value of 67.5% for the external validation set. Additionally, the shapley additive explanation (SHAP) and atom heatmap approaches were utilized to discover the important features and structural fragments related to hematotoxicity, which could offer helpful tips to detect undesired positive substances. Furthermore, matched molecular pair analysis (MMPA) and representative substructure derivation technique were employed to further characterize and investigate the transformation principles and distinctive structural features of hematotoxic chemicals. We believe that the novel graph-based deep learning algorithms and insightful interpretation presented in this study can be used as a trustworthy and effective tool to assess hematotoxicity in the development of new drugs.

Machine Learning Modeling and Insights into the Structural Characteristics of Drug-Induced Neurotoxicity

Neurotoxicity can be resulted from many diverse clinical drugs, which has been a cause of concern to human populations across the world. The detection of drug-induced neurotoxicity (DINeurot) potential with biological experimental methods always required a lot of budget and time. In addition, few studies have addressed the structural characteristics of neurotoxic chemicals. In this study, we focused on the computational modeling for drug-induced neurotoxicity with machine learning methods and the insights into the structural characteristics of neurotoxic chemicals. Based on the clinical drug data with neurotoxicity effects, we developed 35 different classifiers by combining five different machine learning methods and seven fingerprint packages. The best-performing model achieved good results on both 5-fold cross-validation (balanced accuracy of 76.51%, AUC value of 0.83, and MCC value of 0.52) and external validation (balanced accuracy of 83.63%, AUC value of 0.87, and MCC value of 0.67). The model can be freely accessed on the web server DINeuroTpredictor (http://dineurot.sapredictor.cn/). We also analyzed the distribution of several key molecular properties between neurotoxic and non-neurotoxic structures. The results indicated that several physicochemical properties were significantly different between the neurotoxic and non-neurotoxic compounds, including molecular polar surface area (MPSA), AlogP, the number of hydrogen bond acceptors (nHAcc) and donors (nHDon), the number of rotatable bonds (nRotB), and the number of aromatic rings (nAR). In addition, 18 structural alerts responsible for chemical neurotoxicity were identified. The structural alerts have been integrated with our web server SApredictor (http://www.sapredictor.cn). The results of this study could provide useful information for the understanding of the structural characteristics and computational prediction for chemical neurotoxicity.

Molecular Screening and Toxicity Estimation of 260,000 Perfluoroalkyl and Polyfluoroalkyl Substances (PFASs) through Machine Learning

Perfluoroalkyl and polyfluoroalkyl substances (PFASs) are a class of chemicals widely used in industrial applications due to their exceptional properties and stability. However, they do not readily degrade in the environment and are linked to contamination and adverse health effects in humans and wildlife. To find alternatives for the most commonly used PFAS molecules that maintain their desirable chemical properties but are not adverse to biological lifeforms, a novel approach based upon machine learning is utilized. The machine learning model is trained on an existing set of PFAS molecules to generate over 260,000 novel PFAS molecules, which we dub PFAS-AI-Gen. Using molecular descriptors with known relationships to toxicity and industrial suitability followed by molecular docking and molecular dynamics simulations, this set of molecules is screened. In this manner, increasingly complex calculations are performed only for candidate molecules that are most likely to yield the desired properties of low binding affinity toward two selected protein receptors, the human pregnane x receptor (hPXR) and peroxisome proliferator-activated receptor γ (PPAR-γ), and high industrial suitability, defined by critical micelle concentration (CMC). The selection criteria of low binding affinity and high industrial suitability are relative to the popular PFAS alternative GenX. hPXR and PPAR-γ are selected as they are PFAS targets and facilitate a variety of functions, such as drug metabolism and glucose regulation, respectively. Through this approach, 22 promising new PFAS substitutes that may warrant experimental investigation are identified. This integrated approach of molecular screening and toxicity estimation may be applicable to other chemical classes.

Machine learning and artificial intelligence in toxicological sciences

Machine learning and artificial intelligence approaches have revolutionized multiple disciplines, including toxicology. This review summarizes representative recent applications of machine learning and artificial intelligence approaches in different areas of toxicology, including physiologically based pharmacokinetic (PBPK) modeling, quantitative structure-activity relationship modeling for toxicity prediction, adverse outcome pathway analysis, high-throughput screening, toxicogenomics, big data and toxicological databases. By leveraging machine learning and artificial intelligence approaches, now it is possible to develop PBPK models for hundreds of chemicals efficiently, to create in silico models to predict toxicity for a large number of chemicals with similar accuracies compared to in vivo animal experiments, and to analyze a large amount of different types of data (toxicogenomics, high-content image data, etc.) to generate new insights into toxicity mechanisms rapidly, which was impossible by manual approaches in the past. To continue advancing the field of toxicological sciences, several challenges should be considered: (1) not all machine learning models are equally useful for a particular type of toxicology data, and thus it is important to test different methods to determine the optimal approach; (2) current toxicity prediction is mainly on bioactivity classification (yes/no), so additional studies are needed to predict the intensity of effect or dose-response relationship; (3) as more data become available, it is crucial to perform rigorous data quality check and develop infrastructure to store, share, analyze, evaluate, and manage big data; and (4) it is important to convert machine learning models to user-friendly interfaces to facilitate their applications by both computational and bench scientists.