Zoekresultaten

Open for Business: Pathways to strengthen CGIAR's responsible engagement with the private sector

The application of natural language processing for the extraction of mechanistic information in toxicology

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).

Regional variations in childbirth interventions and their correlations with adverse outcomes, birthplace and care provider

Background Variations in childbirth interventions may indicate inappropriate use. Most variation studies are limited by the lack of adjustments for maternal characteristics and do not investigate variations in adverse outcomes. This study aims to explore regional variations in the Netherlands and their correlations with referral rates, birthplace, interventions, and adverse outcomes, adjusted for maternal characteristics. Methods In this nationwide retrospective cohort study, using a national data register, intervention rates were analysed between twelve regions among single childbirths after 37 weeks’ gestation in 2010–2013 (n = 614,730). These were adjusted for maternal characteristics using multivariable logistic regression. Primary outcomes were intrapartum referral, birthplace, and interventions used in midwife- and obstetrician-led care. Correlations both between primary outcomes and between adverse outcomes were calculated with Spearman’s rank correlations. Findings Intrapartum referral rates varied between 55–68% (nulliparous) and 20–32% (multiparous women), with a negative correlation with receiving midwife-led care at the onset of labour in two-thirds of the regions. Regions with higher referral rates had higher rates of severe postpartum haemorrhages. Rates of home birth varied between 6–16% (nulliparous) and 16–31% (multiparous), and was negatively correlated with episiotomy and postpartum oxytocin rates. Among midwife-led births, episiotomy rates varied between 14–42% (nulliparous) and 3–13% (multiparous) and in obstetrician-led births from 46–67% and 14–28% respectively. Rates of postpartum oxytocin varied between 59–88% (nulliparous) and 50–85% (multiparous) and artificial rupture of membranes between 43–52% and 54–61% respectively. A north-south gradient was visible with regard to birthplace, episiotomy, and oxytocin. Conclusions Our study suggests that attitudes towards interventions vary, independent of maternal characteristics. Care providers and policy makers need to be aware of reducing unwarranted variation in birthplace, episiotomy and the postpartum use of oxytocin. Further research is needed to identify explanations and explore ways to reduce unwarranted intervention rates.

Optimization of an adverse outcome pathway network on chemical-induced cholestasis using an artificial intelligence-assisted data collection and confidence level quantification approach

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.

Application of the adverse outcome pathway (AOP) concept to structure the available in vivo and in vitro mechanistic data for allergic sensitization to food proteins

From the publisher: "Background: The introduction of whole new foods in a population may lead to sensitization and food allergy. This constitutes a potential public health problem and a challenge to risk assessors and managers as the existing understanding of the pathophysiological processes and the currently available biological tools for prediction of the risk for food allergy development and the severity of the reaction are not sufficient. There is a substantial body of in vivo and in vitro data describing molecular and cellular events potentially involved in food sensitization. However, these events have not been organized in a sequence of related events that is plausible to result in sensitization, and useful to challenge current hypotheses. The aim of this manuscript was to collect and structure the current mechanistic understanding of sensitization induction to food proteins by applying the concept of adverse outcome pathway (AOP). Main body: The proposed AOP for food sensitization is based on information on molecular and cellular mechanisms and pathways evidenced to be involved in sensitization by food and food proteins and uses the AOPs for chemical skin sensitization and respiratory sensitization induction as templates. Available mechanistic data on protein respiratory sensitization were included to fill out gaps in the understanding of how proteins may affect cells, cell-cell interactions and tissue homeostasis. Analysis revealed several key events (KE) and biomarkers that may have potential use in testing and assessment of proteins for their sensitizing potential. Conclusion: The application of the AOP concept to structure mechanistic in vivo and in vitro knowledge has made it possible to identify a number of methods, each addressing a specific KE, that provide information about the food allergenic potential of new proteins. When applied in the context of an integrated strategy these methods may reduce, if not replace, current animal testing approaches. The proposed AOP will be shared at the www.aopwiki.org platform to expand the mechanistic data, improve the confidence in each of the proposed KE and key event relations (KERs), and allow for the identification of new, or refinement of established KE and KERs." Authors: Jolanda H. M. van BilsenEmail author, Edyta Sienkiewicz-Szłapka, Daniel Lozano-Ojalvo, Linette E. M. Willemsen, Celia M. Antunes, Elena Molina, Joost J. Smit, Barbara Wróblewska, Harry J. Wichers, Edward F. Knol, Gregory S. Ladics, Raymond H. H. Pieters, Sandra Denery-Papini, Yvonne M. Vissers, Simona L. Bavaro, Colette Larré, Kitty C. M. Verhoeckx and Erwin L. Roggen

Natural language processing in toxicology: Delineating adverse outcome pathways and guiding the application of new approach methodologies

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.