Computational Toxicology最新文献

The advancement of protein structural prediction tools, exemplified by AlphaFold and Iterative Threading ASSEmbly Refinement, has enabled the prediction of protein structures across species based on available protein sequence and structural data. In this study, we introduce an innovative molecular docking method that capitalizes on this wealth of structural data to enhance predictions of chemical susceptibility across species. We demonstrated this method using the androgen receptor as a pertinent modulator of endocrine function. By using protein structures, this method contextualizes species susceptibility within a functional framework and helps to integrate molecular docking into the repertoire of New Approach Methodologies (NAMs) that support the Next-Generation Risk Assessment (NGRA) paradigm through the novel integration of various open-source tools.

Genotoxicity assessment is a key component of regulatory decision-making in pesticide authorization and biocide approval. Conventionally, these genotoxicity requirements are addressed with OECD test guideline-compliant in vitro tests. In recent years, in silico approaches, such as (Q)SAR, have matured sufficiently so that they may be suitable to support, complement or even replace in vitro testing as a first tier of genotoxicity assessment. Among the different endpoints for genotoxicity, a high reliability is expected for in silico predictions of the endpoint bacterial mutagenicity. For other endpoints predictive performance is either unclarified or seems to be comparably lower. Herein, we describe the evaluation of several commercial and freely available (Q)SAR models and complementary combinations thereof with respect to the endpoints bacterial mutagenicity and chromosome damage in vitro. We used curated in-house test sets derived from OECD test guideline-compliant studies, gathered from submissions for the regulatory approval of biocides and plant protection products. The data set comprises active substances, metabolites and impurities. In line with previous publications we show that (Q)SAR models for bacterial mutagenicity generally performed well for compounds of the pesticide domain. Model combinations significantly increased the respective sensitivity. Models for chromosome damage still need to improve prior to their stand-alone use in regulatory decision-making, either strongly leaning towards sensitivity, at the expense of specificity or vice versa. Similar to the endpoint bacterial mutagenicity, combinations of models for chromosome damage increase sensitivity when compared to the individual models alone.

Adverse outcome pathway AOPs are conceptual frameworks that organize scientific knowledge about how stressors disrupt specific biological targets, pathways. AOP network consists of two or more AOPs that share common key events (KEs), including crucial events like molecular initiating events (MIEs) and adverse outcomes (AOs), which offer the opportunity to link toxicological pathways. Thus, to better understand the sequential series of KEs involved in the AOP 492 (https://aopwiki.org/aops/492), which is triggered by atrazine (ATZ), we first generated a Reproductive Toxicity via Oxidative Stress (RTOS) AOP network from individual AOPs published in the AOP-Wiki database, using this AOP as a seed. The KEs “Increased, Reactive oxygen species” and “Apoptosis” were considered the most common/highly connected KE within this network and an important point of divergence. Furthermore, “Increased, DNA damage and mutations” is a critical KE within the network, as it is highly connected and central, and represents a point of divergence. This suggests that these three KEs have a high predictive value and could, for example, serve as a basis for the development/selection of in vitro assays to assess reproductive toxicity. The in silico analyses revealed that the pivotal target proteins for ATZ-induced infertility via oxidative stress in humans are Tp53, Bcl2, Esr1, and Nos3, which interact indirectly with ATZ via intermediary factors such as Mapk3, Mapk1, and Cyp19a1. Further, the gene enrichment analyses indicate that these entities are involved in several biological processes and pathways directly associated with oxidative stress, DNA damage and apoptosis, further reinforcing the developed network.

The National Nanotechnology Initiative organized a Nanoinformatics Conference in the 2023 Biden-Harris Administration’s Year of Open Science, which included interested U.S. and EU stakeholders, and preceded the U.S.-EU COR meeting on November 15th, 2023 in Washington, D.C. Progress in the development of a common nanoinformatics infrastructure in the European Union and United States were discussed. Development of contributing, individual database projects, and their strengths and weaknesses, were highlighted. Recommendations and next steps for a U.S. nanoEHS common infrastructure were discussed in light of the pending update of the National Nanotechnology Initiative (NNI)’s Environmental, Health and Safety Research Strategy, and U.S. efforts to curate and house nano Environmental Health and Safety (nanoEHS) data from U.S. federal stakeholder groups. Improved data standards, for reporting and storage have been identified as areas where concerted efforts could most benefit initially. Areas that were not addressed at the conference, but that are critical to progress of the U.S. federal consortium effort are the evaluation of data formats according to use and sustainability measures; modeler and end user, including risk-assessor and regulator perspectives; a need for a community forum or shared data location that is not hosted by any individual U.S. federal agency, and is accessible to the public; as well as emerging needs for integration with new data types such as micro and nano plastics, and interoperability with other data and meta-data, such as adverse outcome pathway information. Future progress will depend on continued interaction of the U.S. and EU CORs, stakeholders and partners in the continued development goals for shared or interoperable infrastructure for nanoEHS.

For organic contaminants in lake sediments, aquifers, and anaerobic bioreactors, their reduction is one of the primary transformation paths in these anoxic water environments. A simple model is introduced to predict pseudo-first order rate constants (k_obs) for the abiotic reduction of organic compounds featuring diverse reducible functional groups. It utilizes the largest experimental dataset of –log k_obs, encompassing 59 organic compounds (278 data points). Unlike available complex quantitative structure–activity relationship (QSAR) methods, the novel approach requires both experimental conditions and structural parameters. In comparison to one of the available general QSAR methods, the new model demonstrates favorable performance. The average absolute deviation (AAD), absolute maximum deviation (AD_max), average absolute relative deviation (AARD%), and R-squared (R²) values of the estimated outputs for 54/5 training/test data sets of the new model are 0.641/1.761, 1.761/1.417, 20.52/83.87, and 0.797/0.949, respectively. On the other hand, the available general comparative QSAR method shows the AAD: 1.311/2.301, AD_max: 3.795/3.732, AARD%: 641.0/821.2, and R²: 0.003/0.447. For the test set, AAD, AARD%, AD_max, and R² values for the new/comparative models are 0.649/2.403, 62.20/190.5, 1.215/3.732 and 0.974/0.789, respectively. In summary, the new model offers a straightforward approach for the manual calculation of –log k_obs, demonstrating excellent goodness-of-fit, reliability, precision, and accuracy.

Justifying read-across predictions for subchronic effects, such as no observed adverse effect levels (NOAEL), is challenging. The scarcity of suitable experimental data hampers such predictions, such that a conservative approach is often employed where the structural similarity between target and the tested source substances is very high. A less stringent interpretation of structural similarity may be used to expand data gap-filling by read-across if other types of similarity (e.g., toxicokinetic and toxicodynamic consideration) are factored into the justification. Herein, qualitative and quantitative in silico-assisted procedures are described and demonstrated for those instances where no structurally similar analogues are identified. In the qualitative approach, the toxicity classification of the most toxic metabolite is assigned directly to the target compound. While simple, this approach may lead to an over-classification of the target compound and a false positive result. In contrast, the quantitative approach is more complicated. In addition to identifying those metabolites causing toxicity, it examines the quantitative information for the amount of the most toxic metabolite. The maximum dose of the parent chemical is estimated which will not result in the generation of toxic metabolites sufficient to cause harmful effects. This quantitative approach permits a calculation of the margin of exposure, is noteworthy for industrial assessment purposes.

MoS-TEC is a newly developed toxicogenomics database for time-expression curves fitted with a statistical model selection approach. Toxicogenomic data provide information on the response of the genome to compounds, often measured in terms of gene expression values. When such experimental data are available for different exposure times, the functional relationships between the exposure time and the expression values of genes might be of interest. The TG-GATEs (Open Toxicogenomics Project-Genomics Assisted Toxicity Evaluation System) database provides such information for genomewide gene expression data for 170 compounds. We performed extensive model selection using MCP-Mod on these data. Specifically, gene expression data measured for eight time points from in vivo experiments on rat liver for 120 compounds with complete datasets were considered. MCP-Mod is a two-step approach, including a multiple comparison procedure (MCP) and a modelling (Mod) approach. The results are estimated time-expression curves that model the relationship between exposure time and gene expression values for all combinations of genes and compounds. We present an appropriate data normalization approach and report which models were selected per compound and in total. For high-quality model fits with a large value for the explained variance, the sigEmax model was most frequently selected. The new R Shiny application MoS-TEC provides easy access for researchers to the best curve fit for all genes individually for all compounds. It can be used online without installing additional software.

The potential toxicity arising from antibiotics and pesticides poses a significant risk to the preservation of groundwater. This study investigates the effects of binary mixtures of pharmaceuticals and pesticides by assessing their log EC₅₀, log EC₃₀, and log EC₁₀ values in relation to Vibrio fischeri bacteria. Based on a comprehensive dataset of 459 observations, this work identifies suitable simple descriptors. Rigorous statistical analysis confirms the models’ reliability, accuracy, precision, and favorable goodness-of-fit. Notably, the ratios of coefficient of determination (R²) for the novel models compared to the best comparative models exceed 1.0: 0.8618/0.8085 for log EC₅₀, 0.8856/0.8422 for log EC₃₀, and 0.8973/0.8556 for log EC₁₀. Additionally, the ratios of root mean square error (RMSE) for the new models relative to their counterparts are all below 1.0: 0.159/0.191 for log EC₅₀, 0.131/0.169 for log EC₃₀, and 0.182/0.215 for log EC₁₀.

Non-animal-based in vitro and in silico approaches for the safety assessment of cosmetic ingredients, recently referred to as Next Generation Risk Assessment (NGRA)/New Approach Methodologies (NAMs), are evolving rapidly as approaches to provide a basis for the regulatory acceptance of new materials. However, predictive models should be applied only to chemicals within the chemical space defined by the dataset used in generating the model. Thus, only predictions for new molecules that are relatively similar to the modeling set can considered reliable with strong confidence. In this study, we developed the S-COPHY model, which employs deep learning to classify new compounds based on their structural similarity to a large collection of pharmaceutical and cosmetic compounds. S-COPHY shows high predictive accuracy both internally and externally, and in particular, there were only a few instances where pharmaceuticals were incorrectly predicted as cosmetics. The use of deep learning enabled the automatic generation of input data from SMILES (Simplified Molecular Input Line Entry System) information, resulting in more consistent model outcomes. Furthermore, GRAD-CAM (Gradient-weighted Class Activation Map) analysis provided insights into the specific structures that contribute to the model's predictions. The potentiality of S-COPHY to identify characteristic structures associated with pharmaceutical-like activity indicates its potential value in supporting safety assessments of cosmetic ingredients. Our results indicate that the S-COPHY model is a promising approach to support decision-making in large chemical spaces, thereby contributing to the safety evaluation of cosmetic ingredients. Expansion of the model to other categories, such as pesticides, could further extend its applicability.

Decision-making on the use and management of chemicals in society is on the brink of a scientific and technological revolution. At the same time world politics is focusing more on chemicals, waste and pollution prevention, alongside climate change and biodiversity loss. To enable effective decision-making, policy makers and regulators will need to draw upon the best scientific evidence available on the real-life causation and consequences of adverse effects of chemical and waste exposures affecting humans, wildlife and the environment. New Approach Method (NAM) data from modern day multidisciplinary science and technology is becoming more available using cheminformatics, computational prediction algorithms using AI, transcriptomics, genomics, proteomics, mathematical modelling, epidemiology, biological monitoring, and clinical science. Current chemical regulation has been shaped by the animal models of the 20th century. NAMs and Next Generation Risk Assessment (NGRA) have the potential to better support innovations in chemicals and materials through science-informed decision making that is more species-relevant and protective of adverse outcomes; this will require future-proofed regulatory transformation. Capacity building and skills development in computational and in vitro NAMs will be key to this transformation.