Comparison of individual and ensemble machine learning models for prediction of sulphate levels in untreated and treated Acid Mine Drainage.

Machine learning was used to provide data for further evaluation of potential extraction of octathiocane (S8), a commercially useful by-product, from Acid Mine Drainage (AMD) by predicting sulphate levels in an AMD water quality dataset. Individual ML regressor models, namely: Linear Regression (LR), Least Absolute Shrinkage and Selection Operator (LASSO), Ridge (RD), Elastic Net (EN), K-Nearest Neighbours (KNN), Support Vector Regression (SVR), Decision Tree (DT), Extreme Gradient Boosting (XGBoost), Random Forest (RF), Multi-Layer Perceptron Artificial Neural Network (MLP) and Stacking Ensemble (SE-ML) combinations of these models were successfully used to predict sulphate levels. A SE-ML regressor trained on untreated AMD which stacked seven of the best-performing individual models and fed them to a LR meta-learner model was found to be the best-performing model with a Mean Squared Error (MSE) of 0.000011, Mean Absolute Error (MAE) of 0.002617 and R2 of 0.9997. Temperature (°C), Total Dissolved Solids (mg/L) and, importantly, iron (mg/L) were highly correlated to sulphate (mg/L) with iron showing a strong positive linear correlation that indicated dissolved products from pyrite oxidation. Ensemble learning (bagging, boosting and stacking) outperformed individual methods due to their combined predictive accuracies. Surprisingly, when comparing SE-ML that combined all models with SE-ML that combined only the best-performing models, there was only a slight difference in model accuracies which indicated that including bad-performing models in the stack had no adverse effect on its predictive performance.

Target and Suspect Screening of Pharmaceuticals and their Transformation Products in the Klip River, South Africa, using Ultra-High-Performance Liquid Chromatography-Mass Spectrometry.

In spite of recent reports about the presence of pharmaceuticals in African water bodies, their prevalence has still not been sufficiently quantified. The few available studies have mostly focused on a limited number of pharmaceuticals. In the present study, a suspect screening of 92 compounds (mainly pharmaceuticals and their transformation products) along the Klip River, South Africa was conducted, followed by target monitoring of 21 of the detected pharmaceuticals. The experimental approach was based on solid-phase extraction followed by analysis with ultra-high-performance liquid chromatography-quadrupole time-of-flight-mass spectrometry (UHPLC-QTOF-MS). The results revealed 47 pharmaceuticals, 31 of which were detected for the first time in South African waters. Seven detected pharmaceuticals (propyphenazole, sulfamerazine, levamisole, tryptophan, dibucaine, albuterol, and fenpropimorph) are not approved medications in South Africa. Six pharmaceutical metabolites were detected for the first time in South Africa. Pharmaceuticals with the highest concentrations in river water were flumequine (0.257 µg L-1 ), oxolinic acid (0.355 µg L-1 ), and acetaminophen (0.432 µg L-1 ). Oxolinic acid presented the highest hazard quotient, 48.6, indicating a risk of toxicity to aquatic organisms. Hazard quotients for other pharmaceuticals were below 1, except that of flumequine, which reached 1.285. These results suggest a need for further research into the fate of pharmaceuticals in surface waters, and a quantification of the risks associated with the identified drugs because they are likely to accumulate in the tissues of fish/aquatic organisms, thus affecting humans. Environ Toxicol Chem 2022;41:437-447. © 2021 SETAC.

A Critical Review of the Antimicrobial and Antibiofilm Activities of Green-Synthesized Plant-Based Metallic Nanoparticles.

Metallic nanoparticles (MNPs) produced by green synthesis using plant extracts have attracted huge interest in the scientific community due to their excellent antibacterial, antifungal and antibiofilm activities. To evaluate these pharmacological properties, several methods or protocols have been successfully developed and implemented. Although these protocols were mostly inspired by the guidelines from national and international regulatory bodies, they suffer from a glaring absence of standardization of the experimental conditions. This situation leads to a lack of reproducibility and comparability of data from different study settings. To minimize these problems, guidelines for the antimicrobial and antibiofilm evaluation of MNPs should be developed by specialists in the field. Being aware of the immensity of the workload and the efforts required to achieve this, we set out to undertake a meticulous literature review of different experimental protocols and laboratory conditions used for the antimicrobial and antibiofilm evaluation of MNPs that could be used as a basis for future guidelines. This review also brings together all the discrepancies resulting from the different experimental designs and emphasizes their impact on the biological activities as well as their interpretation. Finally, the paper proposes a general overview that requires extensive experimental investigations to set the stage for the future development of effective antimicrobial MNPs using green synthesis.

Synthesis and application of cationised cellulose for removal of Cr(VI) from acid mine-drainage contaminated water.

Background: Acid mine drainage (AMD) leads to contamination of surface and ground water by high levels of toxic metals including chromium. In many cases, these waters are sources of drinking water for communities, and treatment is therefore required before consumption to prevent negative health effects. Methods: Cationised hemp cellulose was prepared by etherification with two quaternary ammonium salts: 3-chloro-2-hydroxypropyl trimethyl ammonium chloride (CHPTAC) and glycidyltrimethylammonium chloride (GTMAC) and examined for (i) the efficiency of Cr(VI) removal under acid mine-drainage (AMD) conditions, and (ii) antibacterial activity. Adsorbents were characterised by electron microscopy, Fourier transform infrared (FTIR), CP-MAS 13C nuclear magnetic resonance (NMR) spectroscopy, elemental composition and surface charge. Results: FTIR and solid state 13C NMR confirmed the introduction of quaternary ammonium moieties on cellulose. 13C NMR also showed that cationisation decreased the degree of crystallisation and lateral dimensions of cellulose fibrils. Nevertheless, 47 %  - 72 % of Cr(VI) ions were removed from solutions at pH 4, by 0.1 g of CHPTAC and GTMAC-cationised cellulose, respectively. Adsorption kinetics followed the pseudo-second order model and isotherms were best described by the Freundlich and Dubinin-Radushkevich models. When GTMAC-modified cellulose was applied to AMD contaminated water (pH 2.7); however, Cr(VI) removal decreased to 22% likely due to competition from Al and Fe ions. Cationised materials displayed considerable antibacterial effects, reducing the viability of Escherichia coli by up to 45 % after just 3 hours of exposure. Conclusions: Together, these results suggest that cationised cellulose can be applied in the treatment of Cr(VI)-contaminated mine water particularly if pre-treatments to reduce Fe and Al concentrations are applied.