Using metabolic information for categorization and read-across in the OECD QSAR Toolbox

Toward Metabolic Similarity in Read-Across: A Case Study Using Graph Convolutional Networks to Predict Genotoxicity Outcomes from Simulated Metabolic Networks

Metabolic similarity is a key consideration in evaluating candidate source analogues for read-across (RAx), but approaches to systematically characterize metabolism for read-across prediction are still evolving. Metabolic similarity is multifaceted, considering the similarity of the metabolic tree, the metabolites simulated, and the transformation pathways. The structure of metabolic trees lends itself naturally to graph representations, for which several methods, including graph convolutional networks (GCNs), can be applied to quantify the pairwise similarity between the target and source analogue(s) within an analogue or category approach. In this study, we compared metabolic graph representations of metabolites with structural similarities in predicting genotoxicity outcomes using a data set comprising 5403 chemicals. Xenobiotic metabolism pathways were predicted using the rat liver models within the commercial expert system, TIssue MEtabolism Simulator (TIMES), and the phase I and II xenobiotic metabolism modules within the freely available system BioTransformer. Metabolic pathways were converted to graphs and used to train GCNs, generating embeddings for each chemical. The classification performance of generalized read-across (GenRA), random forest (RF), logistic regression (LR), and multilayer perceptron (MLP) was compared using GCN-derived embeddings versus both Morgan and MACCS chemical fingerprints to identify genotoxic chemicals. GCN embeddings with LR, based on in vivo TIMES metabolism predictions using MACCS fingerprints as node features, achieved the highest area under the curve of the receiver operating characteristic of 0.807, outperforming GenRA and LR with MACCS fingerprints by 14.47% and 5.49%, respectively. Our findings suggest that GCN embeddings of predicted metabolism pathways perform substantially better than structural features of the parent chemicals in predicting genotoxicity outcomes. Such GCN embeddings offer new avenues of systematically encoding end point metabolic information to facilitate analogue identification for read-across.

MetSim: Integrated Programmatic Access and Pathway Management for Xenobiotic Metabolism Simulators

Xenobiotic metabolism is a key consideration in evaluating the hazards and risks posed by environmental chemicals. A number of software tools exist that are capable of simulating metabolites, but each reports its predictions in a different format and with varying levels of detail. This makes comparing the performance and coverage of the tools a practical challenge. To address this shortcoming, we developed a metabolic simulation framework called MetSim, which comprises three main components. A graph-based schema was developed to allow metabolism information to be harmonized. The schema was implemented in MongoDB to store and retrieve metabolic graphs for subsequent analysis. MetSim currently includes an application programming interface for four metabolic simulators: BioTransformer, the OECD Toolbox, EPA's chemical transformation simulator (CTS), and tissue metabolism simulator (TIMES). Lastly, MetSim provides functions to help evaluate simulator performance for specific data sets. In this study, a set of 112 drugs with 432 reported metabolites were compiled, and predictions were made using the 4 simulators. Fifty-nine of the 112 drugs were taken from the Small Molecule Pathway Database, with the remainder sourced from the literature. The human models within BioTransformer and CTS (Phase I only) and the rat models within TIMES and the OECD Toolbox (Phase I only) were used to make predictions for the chemicals in the data set. The recall and precision (recall, precision) ranked in order of highest recall for each individual tool were CTS (0.54, 0.017), BioTransformer (0.50, 0.008), Toolbox in vitro (0.40, 0.144), TIMES in vivo (0.40, 0.133), Toolbox in vivo (0.40, 0.118), and TIMES in vitro (0.39, 0.128). Combining all of the model predictions together increased the overall recall (0.73, 0.008). MetSim enabled insights into the performance and coverage of in silico metabolic simulators to be more efficiently derived, which in turn should aid future efforts to evaluate other data sets.

The prediction of hydrolysis and biodegradation of organophosphorus-based chemical warfare agents (G-series and V-series) using toxicology in silico methods

Nerve agents (G- and V-series) are a group of extremely toxic organophosphorus chemical warfare agents that we have had the opportunity to encounter many times on a massive scale (Matsumoto City, Tokyo subway and Gulf War). The threat of using nerve agents in terrorist attacks or military operations is still present, even with establishing the Chemical Weapons Convention as the legal framework. Understanding their environmental sustainability and health risks is critical to social security. Due to the risk of contact with dangerous nerve agents and animal welfare considerations, in silico methods were used to assess hydrolysis and biodegradation safely. The environmental fate of the examined nerve agents was elucidated using QSAR models. The results indicate that the investigated compounds released into the environment hydrolyse at a different rate, from extremely fast (<1 day) to very slow (over a year); V-agents undergo slower hydrolysis compared to G-agents. V-agents turned out to be relatively challenging to biodegrade, the ultimate biodegradation time frame of which was predicted as weeks to months, while for G-agents, the overwhelming majority was classified as weeks. In silico methods for predicting various parameters are critical to preparing for the forthcoming application of nerve agents.

Artificial intelligence and machine learning disciplines with the potential to improve the nanotoxicology and nanomedicine fields: a comprehensive review

The use of nanomaterials in medicine depends largely on nanotoxicological evaluation in order to ensure safe application on living organisms. Artificial intelligence (AI) and machine learning (MI) can be used to analyze and interpret large amounts of data in the field of toxicology, such as data from toxicological databases and high-content image-based screening data. Physiologically based pharmacokinetic (PBPK) models and nano-quantitative structure-activity relationship (QSAR) models can be used to predict the behavior and toxic effects of nanomaterials, respectively. PBPK and Nano-QSAR are prominent ML tool for harmful event analysis that is used to understand the mechanisms by which chemical compounds can cause toxic effects, while toxicogenomics is the study of the genetic basis of toxic responses in living organisms. Despite the potential of these methods, there are still many challenges and uncertainties that need to be addressed in the field. In this review, we provide an overview of artificial intelligence (AI) and machine learning (ML) techniques in nanomedicine and nanotoxicology to better understand the potential toxic effects of these materials at the nanoscale.

Comparing the performance and coverage of selected in silico (liver) metabolism tools relative to reported studies in the literature to inform analogue selection in read-across: A case study

Changes in the regulatory landscape of chemical safety assessment call for the use of New Approach Methodologies (NAMs) including read-across to fill data gaps. One critical aspect of analogue evaluation is the extent to which target and source analogues are metabolically similar. In this study, a set of 37 structurally diverse chemicals were compiled from the EPA ToxCast inventory to compare and contrast a selection of metabolism in silico tools, in terms of their coverage and performance relative to metabolism information reported in the literature. The aim was to build understanding of the scope and capabilities of these tools and how they could be utilised in a read-across assessment. The tools were Systematic Generation of Metabolites (SyGMa), Meteor Nexus, BioTransformer, Tissue Metabolism Simulator (TIMES), OECD Toolbox, and Chemical Transformation Simulator (CTS). Performance was characterised by sensitivity and precision determined by comparing predictions against literature reported metabolites (from 44 publications). A coverage score was derived to provide a relative quantitative comparison between the tools. Meteor, TIMES, Toolbox, and CTS predictions were run in batch mode, using default settings. SyGMa and BioTransformer were run with user-defined settings, (two passes of phase I and one pass of phase II). Hierarchical clustering revealed high similarity between TIMES and Toolbox. SyGMa had the highest coverage, matching an average of 38.63% of predictions generated by the other tools though was prone to significant overprediction. It generated 5,125 metabolites, which represented 54.67% of all predictions. Precision and sensitivity values ranged from 1.1-29% and 14.7-28.3% respectively. The Toolbox had the highest performance overall. A case study was presented for 3,4-Toluenediamine (3,4-TDA), assessed for the derivation of screening-level Provisional Peer Reviewed Toxicity Values (PPRTVs), was used to demonstrate the practical role in silico metabolism information can play in analogue evaluation as part of a read-across approach.

Selection of Representative Constituents for Unknown, Variable, Complex, or Biological Origin Substance Assessment Based on Hierarchical Clustering

Many of the newly produced and registered substances are complex mixtures or substances of unknown or variable composition, complex reaction products, and biological materials (UVCBs). The latter often consist of a large number of constituents, some of them difficult-to-identify constituents, which complicates their (eco)toxicological assessment. In the present study, through a series of examples, different scenarios for selection of representatives via hierarchical clustering of UVCB constituents are exemplified. Hierarchical clustering allows grouping of the individual chemicals into small sets, where the constituents are similar to each other with respect to more than one criterion. To this end, various similarity criteria and approaches for selection of representatives are developed and analyzed. Two types of selection are addressed: (1) selection of the most "conservative" constituents, which could be also used to support prioritization of UVCBs for evaluation, and (2) obtaining of a small set of chemical representatives that covers the structural and metabolic diversity of the whole target UVCBs or a mixture that can then be evaluated for their environmental and (eco)toxicological properties. The first step is to generate all plausible UVCB or mixture constituents. It was found that the appropriate approach for selecting representative constituents depends on the target endpoint and physicochemical parameters affecting the endpoint of interest. Environ Toxicol Chem 2021;40:3205-3218. © 2021 SETAC.

Automated read-across workflow for predicting acute oral toxicity: I. The decision scheme in the QSAR toolbox

A decision-scheme outlining the steps for identifying the appropriate chemical category and subsequently appropriate tested source analog(s) for data gap filling of a target chemical by read-across is described. The primary features used in the grouping of the target chemical with source analogues within a database of 10,039 discrete organic substances include reactivity mechanisms associated with protein interactions and specific-acute-oral-toxicity-related mechanisms (e.g., mitochondrial uncoupling). Additionally, the grouping of chemicals making use of the in vivo rat metabolic simulator and neutral hydrolysis. Subsequently, a series of structure-based profilers are used to narrow the group to the most similar analogues. The scheme is implemented in the OECD QSAR Toolbox, so it automatically predicts acute oral toxicity as the rat oral LD50 value in log [1/mol/kg]. It was demonstrated that due to the inherent variability in experimental data, classification distribution should be employed as more adequate in comparison to the exact classification. It was proved that the predictions falling in the adjacent GSH categories to the experimentally-stated ones are acceptable given the variation in experimental data. The model performance estimated by adjacent accuracy was found to be 0.89 and 0.54 while based on R². The mechanistic and predictive coverages were >0.85.