
ALTEX, Journal Year: 2024, Volume and Issue: 41(3)
Published: Jan. 1, 2024
Language: Английский
ALTEX, Journal Year: 2024, Volume and Issue: 41(3)
Published: Jan. 1, 2024
Language: Английский
Archives of Toxicology, Journal Year: 2024, Volume and Issue: 98(3), P. 735 - 754
Published: Jan. 20, 2024
The rapid progress of AI impacts diverse scientific disciplines, including toxicology, and has the potential to transform chemical safety evaluation. Toxicology evolved from an empirical science focused on observing apical outcomes exposure, a data-rich field ripe for integration. volume, variety velocity toxicological data legacy studies, literature, high-throughput assays, sensor technologies omics approaches create opportunities but also complexities that can help address. In particular, machine learning is well suited handle integrate large, heterogeneous datasets are both structured unstructured-a key challenge in modern toxicology. methods like deep neural networks, large language models, natural processing have successfully predicted toxicity endpoints, analyzed data, extracted facts generated synthetic data. Beyond automating capture, analysis, prediction, techniques show promise accelerating quantitative risk assessment by providing probabilistic outputs capture uncertainties. enables explanation unravel mechanisms increase trust modeled predictions. However, issues model interpretability, biases, transparency currently limit regulatory endorsement AI. Multidisciplinary collaboration needed ensure development interpretable, robust, human-centered systems. Rather than just human tasks at scale, transformative catalyze innovation how evidence gathered, generated, hypotheses formed tested, performed usher new paradigms assessment. Used judiciously, immense advance toxicology into more predictive, mechanism-based, evidence-integrated discipline better safeguard environmental wellbeing across populations.
Language: Английский
Citations
41Frontiers in Drug Discovery, Journal Year: 2024, Volume and Issue: 4
Published: April 8, 2024
Animals like mice and rats have long been used in medical research to help understand disease test potential new treatments before human trials. However, while animal studies contributed important advances, too much reliance on models can also mislead drug development. This article explains for a general audience how is develop medicines, its benefits limitations, more accurate humane techniques—alternatives testing—could improve this process.
Language: Английский
Citations
20Frontiers in Drug Discovery, Journal Year: 2024, Volume and Issue: 4
Published: Feb. 8, 2024
Machine learning (ML) in toxicological sciences is growing exponentially, which presents unprecedented opportunities and brings up important considerations for using ML this field. This review discusses supervised, unsupervised, reinforcement their applications to toxicology. The application of the scientific method central development a model. These steps involve defining problem, constructing dataset, transforming data feature selection, choosing training model, validation, prediction. need rigorous models becoming more requirement due vast number chemicals interaction with biota. Large datasets make task possible, though selecting databases overlapping chemical spaces, amongst other things, an consideration. Predicting toxicity through machine can have significant societal impacts, including enhancements assessing risks, determining clinical toxicities, evaluating carcinogenic properties, detecting harmful side effects medications. We provide concise overview current state topic, focusing on potential benefits challenges related availability extensive datasets, methodologies analyzing these ethical implications involved applying such models.
Language: Английский
Citations
11Metabolites, Journal Year: 2024, Volume and Issue: 14(2), P. 98 - 98
Published: Jan. 31, 2024
Metabolomics is emerging as a powerful systems biology approach for improving preclinical drug safety assessment. This review discusses current applications and future trends of metabolomics in toxicology development. can elucidate adverse outcome pathways by detecting endogenous biochemical alterations underlying toxicity mechanisms. Furthermore, enables better characterization human environmental exposures their influence on disease pathogenesis. approaches are being increasingly incorporated into studies pharmacology evaluations to gain mechanistic insights identify early biomarkers toxicity. However, realizing the full potential regulatory decision making requires robust demonstration reliability through quality assurance practices, reference materials, interlaboratory studies. Overall, shows great promise strengthening understanding toxicity, enhancing routine screening, transforming exposure risk assessment paradigms. Integration with computational, vitro, personalized medicine innovations will shape predictive toxicology.
Language: Английский
Citations
10ALTEX, Journal Year: 2024, Volume and Issue: unknown
Published: Jan. 1, 2024
Developmental neurotoxicity (DNT) testing has seen enormous progress over the last two decades. Preceding even publication of animal-based OECD test guideline for DNT in 2007, a series non-animal technology workshops and conferences (starting 2005) shaped community that delivered comprehensive battery vitro methods (IVB). Its data interpretation is covered by very recent guidance (No. 377). Here, we aim to overview field, focusing on evolution strategies, role emerging technologies, impact guidelines testing. In particular, this an example targeted development animal-free approach one most complex hazards chemicals human health. These developments started literally from blank slate, with no proposed alternative available. Over decades, cutting-edge science enabled design spares animals enables throughput challenging hazard. While it evident field needs regulation, massive economic decreased cognitive capacity caused chemical exposure should be prioritized more highly. Beyond this, claim fame scientific brought understanding brain, its development, how can perturbed. Plain language summaryDevelopmental predicts hazard brain development. Comprehensive advanced strategies using now replace approaches assess large numbers accurately efficiently than approach. Recent formalized DNT, marking pivotal achievement field. The shift towards reflects both commitment animal welfare growing recognition public health impacts associated impaired function exposures. innovations ultimately contribute safer management better protection health, especially during vulnerable stages
Language: Английский
Citations
10Sustainable Environment, Journal Year: 2025, Volume and Issue: 11(1)
Published: Feb. 13, 2025
Language: Английский
Citations
2Environmental Sciences Europe, Journal Year: 2025, Volume and Issue: 37(1)
Published: March 18, 2025
Language: Английский
Citations
2Frontiers in Toxicology, Journal Year: 2025, Volume and Issue: 7
Published: Feb. 12, 2025
The paradigm shift brought about by artificial intelligence (AI) across scientific disciplines has been nothing short of revolutionary. From unraveling the mysteries protein folding to enabling autonomous systems, AI demonstrated its potential tackle previously intractable problems (Jumper et. al., 2021;Abramoff et 2023). In toxicology, this transformation arrives at a crucial moment, as we face mounting challenges in chemical safety assessment and an urgent need reduce reliance on animal testing (Hartung, 2023a,b;Kleinstreuer Hartung, 2024). This Research Topic emerged from recognition that while computational toxicology made significant strides using classical approaches such physiologically-based pharmacokinetic (PBPK) modeling quantitative structure-activity relationships (QSAR), full modern techniques remains largely untapped toxicological risk assessment. recent advances machine learning, particularly deep natural language processing, semantic interoperability, offer unprecedented opportunities integrate diverse data sources create more predictive models for human health outcomes.The five contributions showcase innovative bridge traditional methods with cutting-edge applications. Collectively, these works demonstrate enhance our understanding hazards advancing development efficient, ethical, human-relevant strategies.Instability" review highlights AI/ML OMICS technologies transform assessments predict genotoxicity mutagenicity higher accuracy. It discusses how can establish biomarkers signatures early cancer detection, assessment, monitoring impacts exposure. Additionally, it emphasizes may accelerate screening chemicals evaluation, optimizing resource use, reducing testing. A common thread throughout is emphasis open science principles reproducibility. authors have shared their code, data, methodologies through public repositories, others build upon work. commitment transparency collaboration exemplifies bringing not only problem-solving but also way conduct share research.Looking ahead, several emerge works. First, shows promise predicting endpoints, integrating predictions into regulatory frameworks hurdle (Hartung Kleinstreuer, 2025).Second, quality accessibility training continue be limiting factors developing robust models. Finally, there pressing standardized validate AI-driven toxicology.Nevertheless, presented are moving closer establishing true probabilistic framework (Maertens 2022(Maertens , 2024) incorporates AI. Such could provide accurate effects.The integration multiple streams -from structures historical vitro assays, omics measurements -through represents promising path forward.We hope will serve both inspiration practical researchers working intersection toxicology. As advance methods, move
Language: Английский
Citations
1ALTEX, Journal Year: 2024, Volume and Issue: 41(1)
Published: Jan. 1, 2024
Green toxicology is marching chemistry into the 21st century. This emerging framework will transform how chemical safety evaluated by incorporating evaluation of hazards, exposures, and risks associated with chemicals early product development in a way that minimizes adverse impacts on human environmental health. The goal to minimize toxic threats across entire supply chains through smarter designs policies. Traditional animal testing methods are replaced faster, cutting-edge innovations like organs-on-chips artificial intelligence predictive models also more cost-effective. Core principles green include utilizing alternative test methods, applying precautionary principle, considering lifetime impacts, emphasizing risk prevention over reaction. paper provides an overview these foundational concepts describes current initiatives future opportunities advance adoption approaches. Challenges limitations discussed. shoots governments offering carrots European Deal nudge industry. Noteworthy, rights environmental groups have different ideas about needs for their consequences use. represents forward support both societal sufficient throughput relevance hazard information minimal suffering. toxicology thus sets stage synergize health ecological values. Overall, integration has potential profoundly shift managed achieve goals ethical, ecologically-conscious manner. Plain language summary aims make safer design. It focuses preventing toxicity issues during instead after products developed. uses modern non-animal computer lab tests cells predict if chemicals could be hazardous. Benefits faster results, lower costs, less testing. using tests, caution even uncertain data, considering global chains, article highlights US policy efforts spur sustainable innovation which necessitate greener approaches assess new materials drive adoption. seeks integrate design so valued equally functionality profit. alignment promises safer, ethical but faces challenges around validating overcoming institutional resistance change.
Language: Английский
Citations
8Public Health Toxicology, Journal Year: 2025, Volume and Issue: 5(1), P. 1 - 3
Published: Feb. 28, 2025
Language: Английский
Citations
1