Non-targeted analysis has emerged as a powerful technique capable of detecting thousands of chemical features in a single environmental sample. However, despite its sophistication, the critical bottleneck remains the confident identification of detected ions. Traditional workflows, which rely heavily on spectral libraries constructed from known compounds, can positively identify only a small percentage of the detected substances. The review underscores that currently less than a few percent of environmentally relevant compounds can be confidently identified using these classical approaches. This limitation curtails the utility of the exhaustive datasets generated and impedes comprehensive environmental monitoring.

Machine learning techniques provide an innovative avenue to surmount this challenge. One of the striking advancements lies in using predictive models that generate tandem mass spectra (MS/MS) from known molecular structures in silico, effectively expanding existing spectral libraries without the need for time-consuming experimental acquisition. These computational spectra act as reference points, greatly enhancing the ability to match unknown spectral signatures to candidate molecules. Moreover, ML models can infer molecular formulas and elucidate fragmentation pathways directly from experimental spectra, narrowing the candidate space significantly, and increasing identification confidence beyond traditional rule-based methods.

The transition from manual, expert-driven interpretation toward automated, scalable analysis is one of the hallmarks of applying machine learning to NTA. The high-dimensional data produced by LC-HRMS poses interpretation challenges that traditional algorithms and heuristics cannot efficiently resolve. Machine learning excels at extracting complex, nonlinear relationships within spectral data, enabling more nuanced and comprehensive insights into unknown chemical signatures. This shift not only expedites pollutant identification but also democratizes access to advanced analytical techniques by reducing reliance on specialized expertise.

Beyond identification, machine learning is harnessed to propose plausible chemical structures de novo through generative modeling approaches. These models, trained on vast chemical databases, are capable of interpreting spectral data to generate candidate molecular structures that may have never been cataloged or previously characterized. This is particularly transformative for emerging contaminants and transformation products—chemical entities arising from environmental processes or industrial activities that evade inclusion in traditional databases. Consequently, researchers can now explore uncharted chemical space with a degree of sophistication previously unattainable.

Integration of orthogonal parameters such as chromatographic retention time and ion mobility collision cross section further elevates the accuracy of structural assignments. Neural network models adept at predicting these properties across different experimental platforms allow for cross-validation of candidate structures, reducing false-positive rates, and instilling greater confidence in identifications. The combination of retention indices, collision cross section predictions, and spectral data constructs a multidimensional targeting framework that enhances pollutant annotation robustness.

Quantifying organic pollutants in environmental samples remains a complex issue, aggravated by the scarcity of authentic reference standards for many compounds. Traditional quantification depends on calibration curves derived from known standards, a luxury unavailable for most non-targeted analytes. Machine learning approaches have addressed this gap by predicting ionization efficiencies and instrumental response factors using molecular descriptors and experimental metadata. These predictive models enable semiquantitative analyses, translating signal intensities into approximate concentrations without experimental standards, thus advancing high-throughput screening with practical environmental relevance.

The ability to accurately quantify pollutants is crucial for exposure assessment, risk analysis, and regulatory decision-making. Machine learning frameworks that predict ionization behaviors facilitate more reliable and standardized quantification across classes of compounds, thereby enabling large-scale environmental surveillance initiatives. This progression represents a significant stride toward translating vast analytical datasets into actionable knowledge for public health protection and ecological management.

Despite impressive advances, several challenges remain before machine learning can be fully integrated into routine environmental pollutant screening. One critical issue is model transferability: predictive performance often declines when algorithms trained on data from one instrument or laboratory are deployed elsewhere, highlighting the need for standardized protocols and universally representative training datasets. Furthermore, training databases currently underrepresent the chemical diversity pertinent to environmental contexts, limiting model generalizability. The interpretability of complex machine learning models also demands improvement to enhance user trust and regulatory acceptance.

To overcome these hurdles, the review advocates for the development of multimodal learning strategies that synthesize molecular features with experimental metadata, including instrument parameters, sample matrix characteristics, and environmental conditions. Such integrative approaches can improve model robustness and adaptability. Moreover, expanding and curating environmental pollutant databases with diverse chemical classes and transformation products will furnish more realistic training sets, fostering better model generalization and precision.

Looking ahead, the authors envision a future where integrated, automated screening platforms powered by machine learning will deliver comprehensive pollutant identification, property prediction, and quantification within a unified framework. By coupling state-of-the-art algorithms with high-resolution analytical instrumentation, these systems would offer real-time, intelligent environmental monitoring solutions capable of handling complex chemical mixtures with unprecedented accuracy and efficiency.

Ultimately, the convergence of machine learning and non-targeted analysis signals a paradigm shift in environmental chemistry. This fusion paves the way for scalable, intelligent screening workflows that can empower researchers and policymakers alike. Enhanced pollutant detection and quantification translate into improved environmental monitoring, better risk assessment, and more informed decision-making processes—cornerstones for safeguarding public health and ecological integrity.

As such, the ongoing integration of artificial intelligence into environmental sciences exemplifies how cutting-edge computational techniques can address longstanding analytical challenges. Ideas once considered distant or infeasible in analytical chemistry become achievable at scale through machine learning’s pattern recognition and predictive capabilities. The progress documented in this review underscores the vast potential of AI-driven strategies to transform our understanding and management of environmental pollutants in a rapidly changing world.

Subject of Research: Not applicable

Article Title: Application of machine learning in non-targeted analysis for environmental organic pollutants

News Publication Date: 10-Feb-2026

References: Liu, Y.-W; Xiong, H.-Y; Liu, J.-H; et al. Application of machine learning in non-targeted analysis for environmental organic pollutants. AI Environ. 2026, 1(1): 11−22. DOI: 10.66178/aie-0026-0003

Image Credits: Liu Yuwei§, Xiong Haoyang§, Liu Jinhua, Xie Huaijun, Chen Jingwen

Keywords: machine learning, non-targeted analysis, environmental pollutants, liquid chromatography, high-resolution mass spectrometry, spectral libraries, molecular identification, quantification, generative models, ionization efficiency, environmental chemistry

Bladder cancer ranks among the most frequently diagnosed cancers worldwide, with NMIBC representing a significant subset of cases characterized by tumor confinement to the bladder’s inner lining without muscle invasion. Conventional cystoscopic examination and urine cytology, though standard, are invasive, expensive, and sometimes inconclusive, especially for low-grade lesions. Hence, the quest for minimally invasive, reliable biomarkers is imperative. Lipids, known for their crucial roles in cellular signaling, membrane structure, and energy homeostasis, have recently been implicated in cancer pathogenesis, opening the door for lipid-based diagnostic strategies.

The recent study harnessed the sophisticated technique of liquid chromatography coupled with high-resolution mass spectrometry (LC-HRMS) to profile plasma lipid metabolites in a cohort of 214 individuals, including 106 NMIBC patients and 108 healthy controls. This technology enables unprecedented resolution and sensitivity in detecting subtle metabolic alterations associated with malignant transformation. By comparing lipidomes between groups, the researchers sought to decipher distinct biochemical signatures indicative of NMIBC presence and progression.

Findings revealed a pronounced disparity in plasma lipid profiles between NMIBC patients and healthy adults, underscoring the metabolic perturbations induced by bladder carcinogenesis. Notably, metabolites such as hydroxy fatty acids, O-linked triacylglycerols (O-TAG), O-linked lysophosphatidylglycerols (O-LPG), and various hydrocarbons were substantially enriched in the NMIBC group. These alterations suggest a profound remodeling of lipid metabolism in cancer cells, possibly reflecting adaptive mechanisms to support rapid proliferation and survival in the tumor microenvironment.

To translate these lipidomic insights into clinical practice, the authors developed predictive models leveraging select lipid panels. One such panel comprising phosphatidylethanolamines PE(14:1/20:0) and PE(18:2/16:0), alongside 19-methyl-heneicosanoic acid, demonstrated robust discriminatory power between NMIBC patients and controls. The model achieved an area under the curve (AUC) of 0.88 in training datasets, and maintained impressive validation performance with an AUC of 0.82. This level of accuracy rivals or surpasses many existing diagnostic modalities, signaling a potential paradigm shift.

Importantly, the model’s diagnostic capability extended effectively to low-grade NMIBC cases, which typically pose greater diagnostic ambiguity. An AUC of 0.81 for this subgroup highlights the panel’s sensitivity in detecting early-stage malignancies, a crucial factor for enabling timely interventions and improving patient prognoses. Additionally, the study explored grading differentiation by constructing a separate lipid panel capable of distinguishing between low- and high-grade NMIBC. This classifier attained an AUC of 0.815, with consistent cross-validation results, affirming its reproducibility and clinical utility.

These revelations support the notion that perturbations in lipid metabolism are not merely epiphenomena but contributory factors in bladder cancer pathophysiology. Lipid alterations may influence membrane fluidity, oxidative stress responses, and oncogenic signaling pathways. Thus, profiling these molecules offers dual benefits: serving as biomarkers for non-invasive diagnosis and providing insights into tumor biology that may guide therapeutic innovations.

Moreover, the use of plasma as a biofluid for lipidomic analysis highlights the feasibility of routine clinical application. Blood samples are relatively easy to obtain and process compared to invasive tissue biopsies, enhancing patient compliance and enabling longitudinal monitoring. Such monitoring could be vital for surveillance post-treatment, detecting recurrences early, and tailoring personalized management strategies based on lipidomic profiles.

The methodological rigor of the study, incorporating 10-fold cross-validation and leave-one-out validation techniques, strengthens the reliability of the results. These statistical approaches mitigate overfitting and affirm the generalizability of lipid biomarker panels across diverse patient populations. As the field advances, further large-scale multi-center studies will be essential to confirm these findings and optimize lipid panels for different demographic groups.

Integrating lipidomic data with other omics platforms, such as genomics and proteomics, could also amplify diagnostic precision and elucidate complex molecular interactions underpinning NMIBC. Systems biology approaches harnessing multi-modal data can enhance biomarker discovery and ultimately foster the development of targeted therapies aimed at lipid metabolism pathways disrupted in bladder cancer.

The study’s implications extend beyond NMIBC, suggesting that plasma lipidomics might be applicable to other urological malignancies and solid tumors where metabolic dysregulation is evident. Broadening this research may uncover universal or cancer-specific lipid signatures, paving the way for universal screening tools or tumor-type tailored diagnostics.

In conclusion, this pioneering research spotlights plasma lipidomics as a formidable approach to identify novel biomarkers capable of diagnosing and grading non-muscle invasive bladder cancer with high accuracy. The identified lipid profiles not only reflect disease presence but correlate with tumor aggressiveness, underscoring their value for early detection and clinical decision-making. As the medical community continues to grapple with the complexities of bladder cancer, lipid metabolite panels represent a promising leap toward more effective, non-invasive, and precise diagnostics fit for the demands of modern oncological practice.

Subject of Research: Non-muscle invasive bladder cancer (NMIBC) diagnosis and grading using plasma lipidomics.

Article Title: Plasma lipidomics for biomarker identification in non-muscle invasive bladder cancer.

Article References:
Zhao, Y., Ji, Z., Sun, W. et al. Plasma lipidomics for biomarker identification in non-muscle invasive bladder cancer. BMC Cancer 25, 1702 (2025). https://doi.org/10.1186/s12885-025-15019-6

Image Credits: Scienmag.com

DOI: 10.1186/s12885-025-15019-6

Xylazine, traditionally employed as a veterinary anesthetic and sedative, has rarely been the focus of human toxicology until recent years. Despite its intended non-human use, its pharmacological properties—primarily its potent sedative and analgesic effects—have unfortunately rendered it attractive as a substance of abuse. The compound’s ability to induce profound central nervous system depression makes it perilous, complicating medical responses to overdoses where xylazine is involved, especially when mixed with opioids or other narcotics. This newfound evidence from Ankara signals a worrying public health challenge and an urgent call to expand toxicological surveillance.

The research leveraged liquid chromatography-high resolution mass spectrometry (LC-HRMS), a cutting-edge analytical technique that offers unmatched sensitivity and specificity for complex biological matrices. By validating this method for detecting xylazine in human blood and urine, the investigators have established a robust protocol for forensic and clinical laboratories worldwide. The validation included meticulous optimization of sample preparation, chromatographic separation, and mass spectrometric detection parameters, ensuring reliable quantitation even at trace concentrations. This methodological advancement allows toxicologists to definitively screen for xylazine, facilitating accurate diagnosis and epidemiological assessments.

Throughout the study, samples were collected from individuals suspected of substance abuse, revealing a previously undocumented pattern of xylazine exposure within the Ankara population. The presence of xylazine was confirmed alongside other psychoactive substances, highlighting its growing popularity as a polysubstance. Intriguingly, this phenomenon matches global trends, where xylazine has increasingly surfaced in illicit drug markets, often clandestinely mixed with opioids such as fentanyl or heroin, thereby heightening overdose risks and complicating clinical interventions.

The detailed pharmacokinetics of xylazine in humans remain insufficiently characterized, but its veterinary data suggest rapid absorption and a relatively short half-life, with metabolism primarily via hepatic cytochrome enzymes. These metabolic pathways produce several derivatives that may also be pharmacologically active or toxic. The study underscores the necessity to identify both the parent compound and its metabolites in biological samples to accurately assess intoxication and drug exposure, which the employed LC-HRMS technique effectively accomplishes.

One of the crucial challenges addressed by the Ankara research team is differentiating xylazine exposure from the myriad of other sedatives and opioids commonly found in forensic cases. The high-resolution mass spectrometry approach provides exact mass measurements that allow clear distinction between structurally similar compounds. This attribute is vital in combating false positives or negatives, which can impede legal proceedings and clinical management. Moreover, the sensitivity achieved by the method ensures detection of even minimal concentrations, capturing early or low-level abuse that might otherwise go unnoticed.

Beyond mere detection, the comprehensive data generated facilitate better understanding of xylazine’s role in intoxication syndromes. The study’s findings contribute valuable information to forensic toxicologists and medical examiners who face increasing incidents of unexplained central nervous system depression, respiratory failure, and even death linked to polysubstance use. By integrating xylazine analysis into routine toxicological panels, frontline clinicians gain an indispensable diagnostic tool to tailor treatment strategies and improve patient outcomes.

Importantly, the revelation from Ankara also emphasizes the dynamic nature of drug abuse markets, which rapidly adapt to regulatory measures by introducing novel or ‘designer’ substances. Xylazine’s veterinary origin initially shielded it from scrutiny, allowing it to infiltrate illegal drug supplies. The research thus advocates for a proactive stance in forensic toxicology, encouraging continuous updating of analytical methodologies to keep pace with emerging threats. This approach can mitigate public health impacts by informing timely interventions and prevention policies.

The broader implications of this study extend beyond Ankara, echoing concerns voiced by international drug monitoring agencies about the rise of synthetic and veterinary-origin drugs in human abuse settings. The findings highlight the necessity for global collaboration between forensic laboratories, healthcare providers, and law enforcement agencies to share intelligence, standardize detection methods, and coordinate responses. Harmonization of analytical standards, as exemplified by this LC-HRMS validation, is key to building a cohesive defense against the ever-evolving drug epidemic.

Furthermore, the study points to an urgent research agenda aimed at elucidating the toxicodynamics of xylazine in humans, including its interactions with other narcotics, dose-response relationships, and long-term consequences of misuse. Clinical trials are impractical due to ethical constraints; thus, observational and forensic data become invaluable. The validated method opens avenues for large-scale epidemiological studies, potentially revealing demographic patterns, geographic hotspots, and temporal trends in xylazine abuse.

In conclusion, this pioneering work from Ankara marks a milestone in forensic toxicology, revealing the hidden specter of xylazine abuse through the application of advanced LC-HRMS techniques. As xylazine continues to pose grave risks in combination with opioids and other substances, this research provides the scientific foundation necessary for improved detection, clinical management, and preventive strategies. It underscores the relentless necessity for vigilance and innovation in addressing modern drug epidemics, safeguarding public health in an era of chemical complexity.

Subject of Research: Detection and validation of xylazine abuse in human blood and urine samples using LC-HRMS.

Article Title: The first report from Ankara on the presence of xylazine abuse in blood and urine samples using a validated LC-HRMS method.

Article References:
Erol Öztürk, Y., Yeter, O., Aslıyüksek, H. et al. The first report from Ankara on the presence of xylazine abuse in blood and urine samples using a validated LC-HRMS method. Int J Legal Med (2025). https://doi.org/10.1007/s00414-025-03562-7

Image Credits: AI Generated