Search

The application of machine learning has recently gained interest from ecotoxicological fields for its ability to model and predict chemical and/or biological processes, such as the prediction of bioconcentration. However, comparison of different models and the prediction of bioconcentration in invertebrates has not been previously evaluated. A comparison of 24 linear and machine learning models is presented herein for the prediction of bioconcentration in fish and important factors that influenced accumulation identified. R2 and root mean square error (RMSE) for the test data (n = 110 cases) ranged from 0.23–0.73 and 0.34–1.20, respectively. Model performance was critically assessed with neural networks and tree-based learners showing the best performance. An optimised 4-layer multi-layer perceptron (14 descriptors) was selected for further testing. The model was applied for cross-species prediction of bioconcentration in a freshwater invertebrate, Gammarus pulex. The model for G. pulex showed good performance with R2 of 0.99 and 0.93 for the verification and test data, respectively. Important molecular descriptors determined to influence bioconcentration were molecular mass (MW), octanol-water distribution coefficient (logD), topological polar surface area (TPSA) and number of nitrogen atoms (nN) among others. Modelling of hazard criteria such as PBT, showed potential to replace the need for animal testing. However, the use of machine learning models in the regulatory context has been minimal to date and is critically discussed herein. The movement away from experimental estimations of accumulation to in silico modelling would enable rapid prioritisation of contaminants that may pose a risk to environmental health and the food chain., Graphical abstract Unlabelled Image, Highlights • Evaluation of 24 models to predict bioconcentration factors in fish is presented. • Machine learning showed good predictive performance. • First machine learning application to predict bioconcentration in invertebrates • Cross-species modelling is limited by case similarity and biological variability. • TPSA, LogD, and Mw were important descriptors for modelling accumulation processes.

Highlights • Blood DNA methylation profiles of 1156 individuals were assessed for age correlation. • Stepwise regression identified 23 age-associated CpG sites in DNA from blood. • A machine learning model based on 16 markers predicted age with a mean error of 3.8 years. • The model predicted age successfully for twins and ‘diseased’ individuals. • A new NGS-based method was combined with machine learning for age prediction., The ability to estimate the age of the donor from recovered biological material at a crime scene can be of substantial value in forensic investigations. Aging can be complex and is associated with various molecular modifications in cells that accumulate over a person’s lifetime including epigenetic patterns. The aim of this study was to use age-specific DNA methylation patterns to generate an accurate model for the prediction of chronological age using data from whole blood. In total, 45 age-associated CpG sites were selected based on their reported age coefficients in a previous extensive study and investigated using publicly available methylation data obtained from 1156 whole blood samples (aged 2–90 years) analysed with Illumina’s genome-wide methylation platforms (27 K/450 K). Applying stepwise regression for variable selection, 23 of these CpG sites were identified that could significantly contribute to age prediction modelling and multiple regression analysis carried out with these markers provided an accurate prediction of age (R2 = 0.92, mean absolute error (MAE) = 4.6 years). However, applying machine learning, and more specifically a generalised regression neural network model, the age prediction significantly improved (R2 = 0.96) with a MAE = 3.3 years for the training set and 4.4 years for a blind test set of 231 cases. The machine learning approach used 16 CpG sites, located in 16 different genomic regions, with the top 3 predictors of age belonged to the genes NHLRC1, SCGN and CSNK1D. The proposed model was further tested using independent cohorts of 53 monozygotic twins (MAE = 7.1 years) and a cohort of 1011 disease state individuals (MAE = 7.2 years). Furthermore, we highlighted the age markers’ potential applicability in samples other than blood by predicting age with similar accuracy in 265 saliva samples (R2 = 0.96) with a MAE = 3.2 years (training set) and 4.0 years (blind test). In an attempt to create a sensitive and accurate age prediction test, a next generation sequencing (NGS)-based method able to quantify the methylation status of the selected 16 CpG sites was developed using the Illumina MiSeq® platform. The method was validated using DNA standards of known methylation levels and the age prediction accuracy has been initially assessed in a set of 46 whole blood samples. Although the resulted prediction accuracy using the NGS data was lower compared to the original model (MAE = 7.5 years), it is expected that future optimization of our strategy to account for technical variation as well as increasing the sample size will improve both the prediction accuracy and reproducibility.

The occurrence of pharmaceutical and illicit drug residues potentially arising from combined sewer overflows (CSOs) in the Central London portion of the Thames Estuary is presented. Approximately 39 million tonnes of untreated sewage enter the River Thames at 57 CSO points annually. Differential analysis of influents and effluents in a major wastewater treatment plant identified seven potential drug-related CSO markers based on removal rates. Three were present in influent at concentrations >1 μg L−1 (caffeine, cocaine and benzoylecgonine). During dry weather, analysis of hourly samples of river water revealed relatively consistent concentrations for most drugs, including CSO markers, over a tidal cycle. River water was monitored over a week in January and July and then daily across six consecutive weeks in November/December 2014. Out of 31 compounds monitored, 27 drug residues were determined in the River Thames and, combined, ranged between ~1000–3500 ng L−1. Total drug concentration generally declined during extended periods of drier weather. For CSO markers, short-term increases in caffeine, cocaine and benzoylecgonine concentration were observed ~24 h after CSO events (especially those occurring at low tide) and generally within one order of magnitude. Timings of elevated occurrence also correlated well with ammonium ion and dissolved oxygen data following CSOs. This work also represents an important study of pharmaceutical occurrence before a major 'super Sewer’ infrastructure upgrade in London aiming to reduce CSOs by 95%.

The effects of pharmaceuticals and personal care products (PPCPs) on aquatic organisms represent a significant current concern. Herein, a targeted metabolomics approach using liquid chromatography-high resolution mass spectrometry (LC-HRMS) is presented to characterise concentration changes in 29 selected metabolites following exposures of aquatic invertebrates, Gammarus pulex, to pharmaceuticals. Method performance revealed excellent linearity (R2 > 0.99), precision (0.1-19%) and lower instrumental limits of detection (0.002-0.20 ng) for all metabolites studied. Three pharmaceuticals were selected representing the low, middle and high range of measured acute measured toxicities (of a total of 26 compounds). Gammarids were exposed to both the no-observed-adverse-effect-level (NOAEL) and the lowest-observed-adverse-effect-level (LOAEL) of triclosan (0.1 and 0.3 mg L-1), nimesulide (0.5 and 1.4 mg L-1) and propranolol (100 and 153 mg L-1) over 24 h. Quantitative metabolite profiling was then performed. Significant changes in metabolite concentrations relative to controls are presented and display distinct clustered trends for each pharmaceutical. Approximately 37% (triclosan), 33% (nimesulide) and 46% (propranolol) of metabolites showed statistically significant time-related effects. Observed changes are also discussed with respect to internal concentrations of the three pharmaceuticals measured using a method based on pulverised liquid extraction, solid phase extraction and LC-MS/MS. Potential metabolic pathways that may be affected by such exposures are also discussed. This represents the first study focussing on quantitative, targeted metabolomics of this lower trophic level benthic invertebrate that may elucidate biomarkers for future risk assessment. © 2016 The Authors., G. McEneff is acknowledged for laboratory assistance and with G. pulex sampling, as well as staff at the Centre for Excellence in Mass Spectrometry at King’s College London for LC-HRMS support. The research leading to these results has received funding from the European Research Council under the European Union's Seventh Framework Programme (FP/2007-2013)/ERC Grant Agreement n. 320737. A JWT Jones Travelling Fellowship was also awarded to C. Gómez-Canela by the Royal Society of Chemistry for his work to be performed at KCL. T. Miller’s collaboration was funded by the Biotechnology and Biological Sciences Research Council scholarship scheme (Reference BB/K501177/1). Funding bodies played no role in the design of the study or decision to publish. The authors declare no financial conflict of interest.

Pharmaceuticals have been considered ‘contaminants of emerging concern’ for more than 20 years. In that time, many laboratory studies have sought to identify hazard and assess risk in the aquatic environment, whilst field studies have searched for targeted candidates and occurrence trends using advanced analytical techniques. However, a lack of a systematic approach to the detection and quantification of pharmaceuticals has provided a fragmented literature of serendipitous approaches. Evaluation of the extent of the risk for the plethora of human and veterinary pharmaceuticals available requires the reliable measurement of trace levels of contaminants across different environmental compartments (water, sediment, biota - of which biota has been largely neglected). The focus on pharmaceutical concentrations in surface waters and other exposure media have therefore limited both the characterisation of the exposome in aquatic wildlife and the understanding of cause and effect relationships. Here, we compile the current analytical approaches and available occurrence and accumulation data in biota to review the current state of research in the field. Our analysis provides evidence in support of the ‘Matthew Effect’ and raises critical questions about the use of targeted analyte lists for biomonitoring. We provide six recommendations to stimulate and improve future research avenues.

Modeling and prediction of polar organic chemical integrative sampler (POCIS) sampling rates (Rs) for 73 compounds using artificial neural networks (ANNs) is presented for the first time. Two models were constructed: the first was developed ab initio using a genetic algorithm (GSD-model) to shortlist 24 descriptors covering constitutional, topological, geometrical and physicochemical properties and the second model was adapted for Rs prediction from a previous chromatographic retention model (RTD-model). Mechanistic evaluation of descriptors showed that models did not require comprehensive a priori information to predict Rs. Average predicted errors for the verification and blind test sets were 0.03 ± 0.02 L d-1 (RTD-model) and 0.03 ± 0.03 L d-1 (GSD-model) relative to experimentally determined Rs. Prediction variability in replicated models was the same or less than for measured Rs. Networks were externally validated using a measured Rs data set of six benzodiazepines. The RTD-model performed best in comparison to the GSD-model for these compounds (average absolute errors of 0.0145 ± 0.008 L d-1 and 0.0437 ± 0.02 L d-1, respectively). Improvements to generalizability of modeling approaches will be reliant on the need for standardized guidelines for Rs measurement. The use of in silico tools for Rs determination represents a more economical approach than laboratory calibrations.

This study considers whether the current standard toxicokinetic methods are an accurate and applicable assessment of xenobiotic exposure in an aquatic freshwater invertebrate. An in vivo exposure examined the uptake and elimination kinetics for eight pharmaceutical compounds in the amphipod crustacean, Gammarus pulex by measuring their concentrations in both biological material and in the exposure medium over a 96 h period. Selected pharmaceuticals included two anti-inflammatories (diclofenac and ibuprofen), two beta-blockers (propranolol and metoprolol), an anti-depressant (imipramine), an anti-histamine (ranitidine) and two beta-agonists (formoterol and terbutaline). Kinetic bioconcentration factors (BCFs) for the selected pharmaceuticals were derived from a first-order one-compartment model using either the simultaneous or sequential modelling methods. Using the simultaneous method for parameter estimation, BCF values ranged from 12 to 212. In contrast, the sequential method for parameter estimation resulted in bioconcentration factors ranging from 19 to 4533. Observed toxicokinetic plots showed statistically significant lack-of-fits and further interrogation of the models revealed a decreasing trend in the uptake rate constant over time for rantidine, diclofenac, imipramine, metoprolol, formoterol and terbutaline. Previous published toxicokinetic data for 14 organic micro-pollutants were also assessed and similar trends were identified to those observed in this study. The decreasing trend of the uptake rate constant over time highlights the need to interpret modelled data more comprehensively to ensure uncertainties associated with uptake and elimination parameters for determining bioconcentration factors are minimised., Graphical abstract, Highlights • Toxicokinetics for eight pharmaceuticals are presented in Gammarus pulex. • Bioconcentration factors ranged from 12 to 4533 and depended on the method used. • Decreasing trends in the uptake (k1) rate constants were observed. • Recognised models and their assumptions may lead to BCF estimation inaccuracies. • Meta-analysis of previous toxicokinetic work revealed similar trends in k1.

This work is licensed under a CC-BY licence - https://creativecommons.org/licenses/by/4.0/ Quantitative measurement of drug consumption biomarkers in wastewater can provide objective information on community drug use patterns and trends. This study presents the measurement of alcohol consumption in 20 cities across 11 countries through the use of wastewater-based epidemiology (WBE), and reports the application of these data for the risk assessment of alcohol on a population scale using the margin of exposure (MOE) approach. Raw 24-h composite wastewater samples were collected over a one-week period from 20 cities following a common protocol. For each sample a specific and stable alcohol consumption biomarker, ethyl sulfate (EtS) was determined by liquid chromatography coupled to tandem mass spectrometry. The EtS concentrations were used for estimation of per capita alcohol consumption in each city, which was further compared with international reports and applied for risk assessment by MOE. The average per capita consumption in 20 cities ranged between 6.4 and 44.3 L/day/1000 inhabitants. An increase in alcohol consumption during the weekend occurred in all cities, however the level of this increase was found to differ. In contrast to conventional data (sales statistics and interviews), WBE revealed geographical differences in the level and pattern of actual alcohol consumption at an inter-city level. All the sampled cities were in the “high risk” category (MOE < 10) and the average MOE for the whole population studied was 2.5. These results allowed direct comparisons of alcohol consumption levels, patterns and risks among the cities. This study shows that WBE can provide timely and complementary information on alcohol use and alcohol associated risks in terms of exposure at the community level. info:eu-repo/semantics/publishedVersion

Background and aims Wastewater‐based epidemiology is an additional indicator of drug use that is gaining reliability to complement the current established panel of indicators. The aims of this study were to: (i) assess spatial and temporal trends of population‐normalized mass loads of benzoylecgonine, amphetamine, methamphetamine and 3,4-methylenedioxymethamphetamine (MDMA) in raw wastewater over 7 years (2011–17); (ii) address overall drug use by estimating the average number of combined doses consumed per day in each city; and (iii) compare these with existing prevalence and seizure data. Design Analysis of daily raw wastewater composite samples collected over 1 week per year from 2011 to 2017. Setting and Participants Catchment areas of 143 wastewater treatment plants in 120 cities in 37 countries. Measurements Parent substances (amphetamine, methamphetamine and MDMA) and the metabolites of cocaine (benzoylecgonine) and of Δ9‐tetrahydrocannabinol (11‐nor‐9‐carboxy‐Δ9‐tetrahydrocannabinol) were measured in wastewater using liquid chromatography–tandem mass spectrometry. Daily mass loads (mg/day) were normalized to catchment population (mg/1000 people/day) and converted to the number of combined doses consumed per day. Spatial differences were assessed world‐wide, and temporal trends were discerned at European level by comparing 2011–13 drug loads versus 2014–17 loads. Findings Benzoylecgonine was the stimulant metabolite detected at higher loads in southern and western Europe, and amphetamine, MDMA and methamphetamine in East and North–Central Europe. In other continents, methamphetamine showed the highest levels in the United States and Australia and benzoylecgonine in South America. During the reporting period, benzoylecgonine loads increased in general across Europe, amphetamine and methamphetamine levels fluctuated and MDMA underwent an intermittent upsurge. Conclusions The analysis of wastewater to quantify drug loads provides near real‐time drug use estimates that globally correspond to prevalence and seizure data.

The recent development of broad-scope high resolution mass spectrometry (HRMS) screening methods has resulted in a much improved capability for new compound identification in environmental samples. However, positive identifications at the ng/L concentration level rely on analytical reference standards for chromatographic retention time (t(R)) and mass spectral comparisons. Chromatographic t(R) prediction can play a role in increasing confidence in suspect screening efforts for new compounds in the environment, especially when standards are not available, but reliable methods are lacking. The current work focuses on the development of artificial neural networks (ANNs) for t(R) prediction in gradient reversed-phase liquid chromatography and applied along with HRMS data to suspect screening of wastewater and environmental surface water samples. Based on a compound t(R) dataset of >500 compounds, an optimized 4-layer back-propagation multi-layer perceptron model enabled predictions for 85% of all compounds to within 2 min of their measured t(R) for training (n = 344) and verification (n = 100) datasets. To evaluate the ANN ability for generalization to new data, the model was further tested using 100 randomly selected compounds and revealed 95% prediction accuracy within the 2-minute elution interval. Given the increasing concern on the presence of drug metabolites and other transformation products (TPs) in the aquatic environment, the model was applied along with HRMS data for preliminary identification of pharmaceutically-related compounds in real samples. Examples of compounds where reference standards were subsequently acquired and later confirmed are also presented. To our knowledge, this work presents for the first time, the successful application of an accurate retention time predictor and HEMS data-mining using the largest number of compounds to preliminarily identify new or emerging contaminants in wastewater and surface waters EU-International Training Network SEWPROF (Marie Curie - PEOPLE Grant): 317205 ; Spanish Ministry of Economy and Competitiveness: CTQ2012-36189 ; Generalitat Valenciana: research group of excellence PROMETEO II/2014/023 ISIC 2012/016; Biotechnology and Biological Sciences Research Council (BBSRC) ; AstraZeneca (under the Global SHE research program): BB/K501177/1