Artificial neural networks (ANNs) could be used in effective drinking water quality management. This review provides an overview about the history of ANNs and their applications and shortcomings in the drinking water sector. From the papers reviewed, it was found that ANNs might be useful modelling tools due to their successful application in areas such as pipes/infrastructure, membrane filtration, coagulation dosage, disinfection residuals, water quality, etc. The most popular ANNs applied were feed-forward networks, especially Multi-layer Perceptrons (MLPs). It was also noted that over the past decade (2006–2016), ANNs have been increasingly applied in the drinking water sector. This, however, is not the case for South Africa where the application of ANNs in distribution systems is little to non-existent. Future research should be directed towards the application of ANNs in South African distribution systems and to develop these models into decision-making tools that water purification facilities could implement.

New advances in sequencing technology and bioinformatics analysis tools have significantly supported the culture-independent analysis of complex microbial communities associated with environmental, plant, animal and human samples. However, previous work has shown that DNA extraction can have a major influence in the community profile. As such there is a constant need for new methods to efficiently and rapidly prepare and analyze DNA for microbiome research, especially in the case new and emerging technology like the Oxford Nanopore Technologies (ONT) MinION. A commercial standard was used, in triplicate, to evaluate three DNA extraction protocols, including two commercially available and one ”in-house” DNA extraction method. All DNA extractions were done as per manufacturer's instructions and prepared with the same commercial ONT 16S sample preparation kit, prior to being analysed using MinION sequencing. Eight MinION 16S datasets of this microbial reference community were obtained. Reads were initially base called and demultiplexed using ONT's Guppy™ sequencing software (version 3.2.4), filtered using NanoFilt and then classified using Usearch. A set of R scripts are presented to process sintax files generated from Usearch and produce an OTU table that can be used for further analyses. All datasets were deposited into the SRA (NCBI) database. These datasets will allow future extraction kit comparisons using MinION sequencing since a standardize laboratory process using commercially available components, such as the MinION 16S sample preparation kit, microbial reference community and extraction kits, were used. The current ONT 16S workflow making use of the Epi2me agent only provides QC metrics and the ID's of the main genera identified and does not provide any tools currently for further downstream community comparison. The analyses scripts provided in the supplementary material will thus further enable the testing of new datasets against these reference sets and provide users the ability to compare their workflows with ours, thus standardizing comparisons and workflows.