To study the ways in which compounds can induce adverse effects, toxicologists have been constructing Adverse Outcome Pathways (AOPs). An AOP can be considered as a pragmatic tool to capture and visualize mechanisms underlying different types of toxicity inflicted by any kind of stressor, and describes the interactions between key entities that lead to the adverse outcome on multiple biological levels of organization. The construction or optimization of an AOP is a labor intensive process, which currently depends on the manual search, collection, reviewing and synthesis of available scientific literature. This process could however be largely facilitated using Natural Language Processing (NLP) to extract information contained in scientific literature in a systematic, objective, and rapid manner that would lead to greater accuracy and reproducibility. This would support researchers to invest their expertise in the substantive assessment of the AOPs by replacing the time spent on evidence gathering by a critical review of the data extracted by NLP. As case examples, we selected two frequent adversities observed in the liver: namely, cholestasis and steatosis denoting accumulation of bile and lipid, respectively. We used deep learning language models to recognize entities of interest in text and establish causal relationships between them. We demonstrate how an NLP pipeline combining Named Entity Recognition and a simple rules-based relationship extraction model helps screen compounds related to liver adversities in the literature, but also extract mechanistic information for how such adversities develop, from the molecular to the organismal level. Finally, we provide some perspectives opened by the recent progress in Large Language Models and how these could be used in the future. We propose this work brings two main contributions: 1) a proof-of-concept that NLP can support the extraction of information from text for modern toxicology and 2) a template open-source model for recognition of toxicological entities and extraction of their relationships. All resources are openly accessible via GitHub (https://github.com/ontox-project/en-tox).
DOCUMENT
Adverse Outcome Pathways (AOPs) are conceptual frameworks that tie an initial perturbation (molecular initiat- ing event) to a phenotypic toxicological manifestation (adverse outcome), through a series of steps (key events). They provide therefore a standardized way to map and organize toxicological mechanistic information. As such, AOPs inform on key events underlying toxicity, thus supporting the development of New Approach Methodologies (NAMs), which aim to reduce the use of animal testing for toxicology purposes. However, the establishment of a novel AOP relies on the gathering of multiple streams of evidence and infor- mation, from available literature to knowledge databases. Often, this information is in the form of free text, also called unstructured text, which is not immediately digestible by a computer. This information is thus both tedious and increasingly time-consuming to process manually with the growing volume of data available. The advance- ment of machine learning provides alternative solutions to this challenge. To extract and organize information from relevant sources, it seems valuable to employ deep learning Natural Language Processing techniques. We review here some of the recent progress in the NLP field, and show how these techniques have already demonstrated value in the biomedical and toxicology areas. We also propose an approach to efficiently and reliably extract and combine relevant toxicological information from text. This data can be used to map underlying mechanisms that lead to toxicological effects and start building quantitative models, in particular AOPs, ultimately allowing animal-free human-based hazard and risk assessment.
DOCUMENT
Background: Adverse outcome pathway (AOP) networks are versatile tools in toxicology and risk assessment that capture and visualize mechanisms driving toxicity originating from various data sources. They share a common structure consisting of a set of molecular initiating events and key events, connected by key event relationships, leading to the actual adverse outcome. AOP networks are to be considered living documents that should be frequently updated by feeding in new data. Such iterative optimization exercises are typically done manually, which not only is a time-consuming effort, but also bears the risk of overlooking critical data. The present study introduces a novel approach for AOP network optimization of a previously published AOP network on chemical-induced cholestasis using artificial intelligence to facilitate automated data collection followed by subsequent quantitative confidence assessment of molecular initiating events, key events, and key event relationships. Methods: Artificial intelligence-assisted data collection was performed by means of the free web platform Sysrev. Confidence levels of the tailored Bradford-Hill criteria were quantified for the purpose of weight-of-evidence assessment of the optimized AOP network. Scores were calculated for biological plausibility, empirical evidence, and essentiality, and were integrated into a total key event relationship confidence value. The optimized AOP network was visualized using Cytoscape with the node size representing the incidence of the key event and the edge size indicating the total confidence in the key event relationship. Results: This resulted in the identification of 38 and 135 unique key events and key event relationships, respectively. Transporter changes was the key event with the highest incidence, and formed the most confident key event relationship with the adverse outcome, cholestasis. Other important key events present in the AOP network include: nuclear receptor changes, intracellular bile acid accumulation, bile acid synthesis changes, oxidative stress, inflammation and apoptosis. Conclusions: This process led to the creation of an extensively informative AOP network focused on chemical-induced cholestasis. This optimized AOP network may serve as a mechanistic compass for the development of a battery of in vitro assays to reliably predict chemical-induced cholestatic injury.
DOCUMENT
Both because of the shortcomings of existing risk assessment methodologies, as well as newly available tools to predict hazard and risk with machine learning approaches, there has been an emerging emphasis on probabilistic risk assessment. Increasingly sophisticated AI models can be applied to a plethora of exposure and hazard data to obtain not only predictions for particular endpoints but also to estimate the uncertainty of the risk assessment outcome. This provides the basis for a shift from deterministic to more probabilistic approaches but comes at the cost of an increased complexity of the process as it requires more resources and human expertise. There are still challenges to overcome before a probabilistic paradigm is fully embraced by regulators. Based on an earlier white paper (Maertens et al., 2022), a workshop discussed the prospects, challenges and path forward for implementing such AI-based probabilistic hazard assessment. Moving forward, we will see the transition from categorized into probabilistic and dose-dependent hazard outcomes, the application of internal thresholds of toxicological concern for data-poor substances, the acknowledgement of user-friendly open-source software, a rise in the expertise of toxicologists required to understand and interpret artificial intelligence models, and the honest communication of uncertainty in risk assessment to the public.
DOCUMENT
Implementation of reliable methodologies allowing Reduction, Refinement, and Replacement (3Rs) of animal testing is a process that takes several decades and is still not complete. Reliable methods are essential for regulatory hazard assessment of chemicals where differences in test protocol can influence the test outcomes and thus affect the confidence in the predictive value of the organisms used as an alternative for mammals. Although test guidelines are common for mammalian studies, they are scarce for non-vertebrate organisms that would allow for the 3Rs of animal testing. Here, we present a set of 30 reporting criteria as the basis for such a guideline for Developmental and Reproductive Toxicology (DART) testing in the nematode Caenorhabditis elegans. Small organisms like C. elegans are upcoming in new approach methodologies for hazard assessment; thus, reliable and robust test protocols are urgently needed. A literature assessment of the fulfilment of the reporting criteria demonstrates that although studies describe methodological details, essential information such as compound purity and lot/batch number or type of container is often not reported. The formulated set of reporting criteria for C. elegans testing can be used by (i) researchers to describe essential experimental details (ii) data scientists that aggregate information to assess data quality and include data in aggregated databases (iii) regulators to assess study data for inclusion in regulatory hazard assessment of chemicals.
DOCUMENT
Editorial on the Research Topic "Leveraging artificial intelligence and open science for toxicological risk assessment"
LINK
For almost fifteen years, the availability and regulatory acceptance of new approach methodologies (NAMs) to assess the absorption, distribution, metabolism and excretion (ADME/biokinetics) in chemical risk evaluations are a bottleneck. To enhance the field, a team of 24 experts from science, industry, and regulatory bodies, including new generation toxicologists, met at the Lorentz Centre in Leiden, The Netherlands. A range of possibilities for the use of NAMs for biokinetics in risk evaluations were formulated (for example to define species differences and human variation or to perform quantitative in vitro-in vivo extrapolations). To increase the regulatory use and acceptance of NAMs for biokinetics for these ADME considerations within risk evaluations, the development of test guidelines (protocols) and of overarching guidance documents is considered a critical step. To this end, a need for an expert group on biokinetics within the Organisation of Economic Cooperation and Development (OECD) to supervise this process was formulated. The workshop discussions revealed that method development is still required, particularly to adequately capture transporter mediated processes as well as to obtain cell models that reflect the physiology and kinetic characteristics of relevant organs. Developments in the fields of stem cells, organoids and organ-on-a-chip models provide promising tools to meet these research needs in the future.
DOCUMENT
Summary: Xpaths is a collection of algorithms that allow for the prediction of compound-induced molecular mechanisms of action by integrating phenotypic endpoints of different species; and proposes follow-up tests for model organisms to validate these pathway predictions. The Xpaths algorithms are applied to predict developmental and reproductive toxicity (DART) and implemented into an in silico platform, called DARTpaths.
DOCUMENT
Despite changing attitudes towards animal testing and current legislation to protect experimental animals, the rate of animal experiments seems to have changed little in recent years. On May 15–16, 2013, the In Vitro Testing Industrial Platform (IVTIP) held an open meeting to discuss the state of the art in alternative methods, how companies have, can, and will need to adapt and what drives and hinders regulatory acceptance and use. Several key messages arose from the meeting. First, industry and regulatory bodies should not wait for complete suites of alternative tests to become available, but should begin working with methods available right now (e.g., mining of existing animal data to direct future studies, implementation of alternative tests wherever scientifically valid rather than continuing to rely on animal tests) in non-animal and animal integrated strategies to reduce the numbers of animals tested. Sharing of information (communication), harmonization and standardization (coordination), commitment and collaboration are all required to improve the quality and speed of validation, acceptance, and implementation of tests. Finally, we consider how alternative methods can be used in research and development before formal implementation in regulations. Here we present the conclusions on what can be done already and suggest some solutions and strategies for the future.
DOCUMENT
Development of novel testing strategies to detect adverse human health effects is of interest to replace in vivo-based drug and chemical safety testing. The aim of the present study was to investigate whether physiologically based kinetic (PBK) modeling-facilitated conversion of in vitro toxicity data is an adequate approach to predict in vivo cardiotoxicity in humans. To enable evaluation of predictions made, methadone was selected as the model compound, being a compound for which data on both kinetics and cardiotoxicity in humans are available. A PBK model for methadone in humans was developed and evaluated against available kinetic data presenting an adequate match. Use of the developed PBK model to convert concentration–response curves for the effect of methadone on human-induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CM) in the so-called multi electrode array (MEA) assay resulted in predictions for in vivo dose–response curves for methadone-induced cardiotoxicity that matched the available in vivo data. The results also revealed differences in protein plasma binding of methadone to be a potential factor underlying variation between individuals with respect to sensitivity towards the cardiotoxic effects of methadone. The present study provides a proof-of-principle of using PBK modeling-based reverse dosimetry of in vitro data for the prediction of cardiotoxicity in humans, providing a novel testing strategy in cardiac safety studies.
DOCUMENT
