Advancing non-target analysis of emerging environmental contaminants with machine learning: Current status and future implications

Environment International, Год журнала: 2025, Номер unknown, С. 109404 - 109404

Опубликована: Март 1, 2025

Emerging environmental contaminants (EECs) such as pharmaceuticals, pesticides, and industrial chemicals pose significant challenges for detection identification due to their structural diversity lack of analytical standards. Traditional targeted screening methods often fail detect these compounds, making non-target analysis (NTA) using high-resolution mass spectrometry (HRMS) essential identifying unknown or suspected contaminants. However, interpreting the vast datasets generated by HRMS is complex requires advanced data processing techniques. Recent advancements in machine learning (ML) models offer great potential enhancing NTA applications. As such, we reviewed key developments, including optimizing workflows computational tools, improved chemical structure identification, quantification methods, enhanced toxicity prediction capabilities. It also discusses future perspectives field, refining ML tools mixtures, improving inter-laboratory validation, further integrating into risk assessment frameworks. By addressing challenges, ML-assisted can significantly enhance detection, quantification, evaluation EECs, ultimately contributing more effective monitoring public health protection.

Язык: Английский

---

Prioritization Strategies for Non-Target Screening in Environmental Samples by Chromatography – High-Resolution Mass Spectrometry: A Tutorial

Journal of Chromatography A, Год журнала: 2025, Номер unknown, С. 465944 - 465944

Опубликована: Апрель 1, 2025

Язык: Английский

---

FT-GNN Tool for Bridging HRMS Features and Bioactivity: Uncovering Unidentified Estrogen Receptor Agonists in Sewage

Environmental Science & Technology, Год журнала: 2025, Номер unknown

Опубликована: Апрель 9, 2025

Identifying primary estrogen receptor (ER) agonists in municipal sewage is essential for ensuring the health of aquatic environments. Given complex and variable chemical composition sewage, predominant ER remain unclear. High-resolution mass spectrometry (HRMS)-based models have been developed to predict compound bioactivity matrices, but further optimization needed effectively bridge HRMS features with agonists. To address this challenge, an FT-GNN (fragmentation tree-based graph neural network) model was proposed. limited data class imbalance, augmentation performed using predictions within applicability domain (AD) oversampling technique (OTE). Model development results demonstrated that integrating improved balanced accuracy (bACC) value by 6%-31%. The model, a high bACC identify more true agonists, efficiently classified tens thousands unidentified reducing postprocessing workload nontargeted screening. Analysis agonist transformation during treatment revealed anaerobic stage as key both their removal formation. Estrogenic effect balance analysis suggests α-E2 9,11-didehydroestriol may be two previously overlooked Collectively, application are crucial advancements toward credible tracking efficient control estrogenic risks water.

Язык: Английский

---

Chemometrical assessment of adverse effects in lung cells induced by vehicle engine emissions

Nanotoxicology, Год журнала: 2025, Номер unknown, С. 1 - 14

Опубликована: Апрель 9, 2025

Vehicle engine exhausts contain complex mixtures of gaseous and particulate pollutants, which are known to affect lung functions adversely. Many in vitro studies have shown that exposure exhaust can induce oxidative stress cells, leading cellular inflammation cytotoxicity. However, it remains challenging identify key harmful components their specific adverse effects via traditional toxicological assessments. Machine learning (ML) methods offer new ways analyzing such datasets gained attention predicting toxicity outcomes identifying pollutants responsible for a non-biased way. This study aims understand the contribution cell using ML techniques. Data were reanalyzed from previous (2015-2018), where 3D human epithelial airway tissue model was exposed gasoline diesel under air-liquid interface (ALI) conditions with different fuels after-treatment systems. dataset included characteristics (particle number (PN), carbon monoxide (CO), total hydrocarbons (THC), nitrogen oxides (NOx) levels) corresponding biological responses (cytotoxicity, stress, inflammatory responses). The relationships between explored techniques, including hierarchical nonhierarchical clustering principal component analysis. findings reveal both (CO, THC, NOx) contribute inflammation, cytotoxicity highlighting significant role each component. In addition, unmeasured factors beyond CO, NOx, PN likely effects, indicating need more detailed characterization parameters By successfully integrating this shows potential pollutant-specific contributions toxicity. These insights guide analysis scenarios inform regulatory measures technical developments emission control.

Язык: Английский

---

Prioritization of Unknown LC-HRMS Features Based on Predicted Toxicity Categories

Environmental Science & Technology, Год журнала: 2025, Номер unknown

Опубликована: Апрель 20, 2025

Complex environmental samples contain a diverse array of known and unknown constituents. While liquid chromatography coupled with high-resolution mass spectrometry (LC-HRMS) nontargeted analysis (NTA) has emerged as an essential tool for the comprehensive study such samples, identification individual constituents remains significant challenge, primarily due to vast number detected features in each sample. To address this, prioritization strategies are frequently employed narrow focus most relevant further analysis. In this study, we developed novel strategy that directly links fragmentation chromatographic data aquatic toxicity categories, bypassing need compounds. Given not always well-characterized through fragmentation, created two models: (1) Random Forest Classification (RFC) model, which classifies fish categories based on MS1, retention, data─expressed cumulative neutral losses (CNLs)─when information is available, (2) Kernel Density Estimation (KDE) model relies solely retention time MS1 when absent. Both models demonstrated accuracy comparable structure-based prediction methods. We tested pesticide mixture tea extract measured by LC-HRMS, where CNL-based RFC achieved 0.76 KDE reached 0.61, showcasing their robust performance real-world applications.

Язык: Английский

---