ALAlison S. Low

Alison S. Low

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Alison S. Low has a total of 40 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Benchmarking nanopore-based strategies for antimicrobial resistance prediction

    by Natalie Ring, Alison S. Low, Rhodri Evans, Marianne Keith, Gavin K. Paterson, David Gally, Tim Nuttall, Dylan Neil Clements, J. Ross Fitzgerald

    Antimicrobial resistance (AMR) presents a pressing need to ensure that the right antimicrobials are used to target the right microbes at the right time. Ideally, the appropriate antimicrobial is selected after patient samples have been cultured and assessed with antimicrobial sensitivity testing (AST). However, the time needed for culture-based diagnosis leads to immediate empirical treatment, often with broad-spectrum and/or high-tier antimicrobials. Direct nanopore metagenomic whole-genome sequencing to identify pathogens and predict their AMR is a rapid and patient-side alternative. A limitation of this approach is potential inconsistencies in i predicted AMR phenotypes. Here, we benchmarked the current performance of in silico AMR prediction strategies for nanopore-generated long-read data. Using real-world nanopore data paired with AST phenotyping for 201 samples representing 27 bacterial species, we assessed the impact of basecalling mode, data volume and assembly strategy and compared the performance of eight in silico AMR prediction tools with seven AMR databases. We found that basecalling accuracy mode does not significantly affect the overall accuracy of in silico AMR predictions but assembly strategy and data volume both do. Prediction tools using the ResFinder database scored best for balanced accuracy (0.80±0.02 for both ResFinder and ABRicate), whilst DeepARG scored best for sensitivity (0.65±0.03); predictions were more accurate for some antibiotic classes and genera than others. However, even the best-performing in silico AMR prediction strategy missed some resistance identified by lab-based AST. We conclude, therefore, that currently in silico AMR prediction can supplement lab-based AST but cannot yet replace it.

  • Open Access

    Analysis of a detoxified Escherichia coli strain for bacteriophage production

    by Emily A. Welham, Alba Park de la Torriente, Jia Arng Lee, Marianne Keith, Sean P McAteer, Gavin K. Paterson, David L. Gally, Alison S. Low

    Phage therapeutics are re-emerging as adjuncts or alternatives to antibiotics and their clinical translation will be enhanced with production methods that minimise downstream processing. We evaluated whether an endotoxin-reduced E. coli strain developed for production of recombinant proteins, ClearColi®, can serve as a useful, safe phage production host without compromising yield and whether targeted receptor complementation can expand its utility. The parent strain BL21(DE3), and its lipid A modified derivative, ClearColi®, were compared with respect to infection and generation of phage. Across a panel of 31 phage, a similar host range was observed between BL21(DE3) and ClearColi®. To expand host range ompC was genetically engineered into the chromosome of ClearColi®, thereby adding OmpC-dependent phage to its production capacity. Production metrics were broadly comparable between the hosts; efficiency of plating and final titres for representative phage were not significantly different; burst size varied by phage but without consistent host bias. Endotoxin activity in ClearColi®-propagated lysates was reduced by over 1000-fold relative to BL21(DE3), reaching the low hundreds of endotoxin units (EU) versus hundreds of thousands for BL21(DE3). Intravesical administration of ClearColi®-derived phage (LUC4) into pigs elicited no clinical abnormalities and no significant increases in circulating cytokines up to 48 hours after administration. ClearColi® allows efficient production of diverse phage with low endotoxin, reducing the requirement for downstream processing. Although its minimal LPS reduces its capacity for producing some LPS-dependent phage and its growth is slower than BL21(DE3), requiring optimisation for maximal phage titre, the safety and simplified manufacturing process support further development of endotoxin modified strains for phage production. Impact statement: Antibiotic resistance is a current global problem and treatments based on phage and phage products already have a proven track record with particular bacterial infections, especially in the urinary tract. While progress is being made on in vitro phage synthesis, large scale bacteriophage preparations require a bacterial host for production, consequently toxic components in the initial lysate need to be removed or significantly diluted for safe clinical use. This is a study of the potential to utilise an endotoxin-reduced E. coli strain, ClearColi®, to produce safer phage therapeutics. Such endotoxin modified strains should minimise the processing steps required and reduce overall production costs of a phage preparation. The research demonstrates that the endotoxin-reduced strain was able to produce a wide range of phage and for studied examples at phage titres equivalent to the more toxic parent strain. We also show that the strain can be modified to increase its host range and confirm the very low endotoxicity of basic phage lysates produced by the strain. Replicating this process to engineer additional low-toxicity bacterial production strains will accelerate the development of safer, more cost-effective phage therapeutics.

  • Open Access

    Quorum sensing and capsule expression enable subpopulation evasion of phage killing in Escherichia coli ST131: Implications for targeted therapy

    by Alba Park de la Torriente, Talal Hossain, Sean P. McAteer, Marianne Keith, Gavin K. Paterson, Alison S. Low, David L. Gally

    Escherichia coli ST131 is a globally disseminated multidrug-resistant lineage frequently associated with recalcitrant urinary tract infections (UTIs) and bacteraemia. While bacteriophages offer a promising alternative treatment to antibiotics, their efficacy is often limited in physiologically relevant conditions in comparison to laboratory media. In this study, we have investigated the mechanisms by which the representative ST131 strain, EC958, evades elimination by a model phage, LUC4. We observed that in the urine environment, EC958 can transiently resist phage infection by a density dependent mechanism and by the production of protective polysaccharides. Based on this understanding, we developed a phage treatment strategy that can sterilise an EC958 culture in urine-based medium, even at high bacterial densities. The rational design of the successful phage therapy strategy utilises a tailored phage cocktail containing phage that encode depolymerase enzymes to degrade bacterial surface carbohydrates and the targeting of multiple receptors to prevent the emergence of fixed genetic resistant mutants. We found the addition of specific carbon sources renders the bacteria more susceptible to phage infection. By combining these findings with a simulated bladder wash to model voiding, we successfully achieved elimination of EC958 cultures in a urine environment. This study provides a framework for overcoming both fixed and reversible phage resistance, offering a translatable strategy for effectively treating urinary tract infections with phage. Author Summary: In this study, we investigated how bacterial populations can overcome a phage infection. Phage are viruses that naturally kill bacteria and provide an alternative treatment to antibiotics. We focussed on a particularly aggressive and antibiotic resistant strain of E. coli , EC958, which belongs to a group of E. coli strains that are a leading cause of urinary tract infections and life-threatening bloodstream infections worldwide. We found that in a simulated bladder environment, these bacteria do not rely on genetic mutations to survive but they employ a range of ‘hide and seek’ strategies. We showed that bacteria can coat themselves in a protective layer to block the phage and use social signalling to enter a dormant state when cell density is high. When they are in this sleep-like state the phage cannot successfully infect. To overcome these bacterial defences, we developed a treatment strategy combining effective phage with specific naturally occurring additives, that trick the bacteria into waking up and becoming vulnerable again to phage infection. By also simulating a clinical bladder wash to reduce bacterial numbers and therefore reduce social-signalling, we were able to eliminate the bacterial population. Our findings suggest that by understanding bacterial strategies we can design more effective and personalised phage therapies to treat bacterial infections.

  • Open Access

    A systematic comparison of tools for predicting antimicrobial resistance from nanopore sequence data

    by Natalie Ring, Alison S Low, Rhodri Evans, Marianne Keith, Gavin K. Paterson, David Gally, Tim Nuttall, Dylan Neil Clements, J. Ross Fitzgerald

    Antimicrobial resistance (AMR) presents a pressing need to ensure that the right antimicrobials are used to target the right microbes at the right time. Ideally, the appropriate antimicrobial is selected after patient samples have been cultured and assessed with antimicrobial sensitivity testing (AST). However, the time needed for culture-based diagnosis leads to immediate empirical treatment, often with broad-spectrum and/or high-tier antimicrobials. Direct nanopore metagenomic whole genome sequencing to identify the pathogens and predict their antimicrobial resistance is a rapid and patient-side alternative. A limitation of this approach is the inconsistency of in silico predicted AMR phenotypes. Here, we benchmarked the current performance of in silico AMR prediction strategies for nanopore-generated long read data. Using nanopore data paired with AST phenotyping for 201 samples, we assessed the impact of basecalling mode, data volume, and assembly strategy to compare the performance of eight in silico AMR prediction tools with seven AMR databases. We found that basecalling accuracy mode does not affect the overall accuracy of in silico AMR predictions, but assembly strategy and data volume both do. Prediction tools using the ResFinder database scored best for balanced accuracy (0.8 for both ResFinder and ABRicate), whilst DeepARG scored best for sensitivity (0.65). However, even the best performing in silico AMR prediction strategy missed some resistance identified by lab-based AST. In silico AMR prediction can therefore supplement lab-based AST, but cannot yet replace it.

  • Open Access

    Predictive phage therapy for Escherichia coli urinary tract infections: Cocktail selection for therapy based on machine learning models

    by Marianne Keith, Alba Park de la Torriente, Antonia Chalka, Adriana Vallejo-Trujillo, Sean P. McAteer, Gavin K. Paterson, Alison S. Low, David L. Gally

    This study supports the development of predictive bacteriophage (phage) therapy: the concept of phage cocktail selection to treat a bacterial infection based on machine learning (ML) models. For this purpose, ML models were trained on thousands of measured interactions between a panel of phage and sequenced bacterial isolates. The concept was applied to Escherichia coli associated with urinary tract infections. This is an important common infection in humans and companion animals from which multidrug-resistant (MDR) bloodstream infections can originate. The global threat of MDR infection has reinvigorated international efforts into alternatives to antibiotics including phage therapy. E. coli exhibit extensive genome-level variation due to horizontal gene transfer via phage and plasmids. Associated with this, phage selection for E. coli is difficult as individual isolates can exhibit considerable variation in phage susceptibility due to differences in factors important to phage infection including phage receptor profiles and resistance mechanisms. The activity of 31 phage was measured on 314 isolates with growth curves in artificial urine. Random Forest models were built for each phage from bacterial genome features, and the more generalist phage, acting on over 20% of the bacterial population, exhibited F1 scores of >0.6 and could be used to predict phage cocktails effective against previously untested strains. The study demonstrates the potential of predictive ML models which integrate bacterial genomics with phage activity datasets allowing their use on data derived from direct sequencing of clinical samples to inform rapid and effective phage therapy.


  • Open Access

    Predictive phage therapy for Escherichia coli urinary tract infections: cocktail selection for therapy based on machine learning models

    by Marianne Keith, Alba Park de la Torriente, Antonia Chalka, Adriana Vallejo-Trujillo, Sean P. McAteer, Gavin K. Paterson, Alison S. Low, David L. Gally

    This study supports the development of predictive bacteriophage (phage) therapy: the concept of phage cocktail selection to treat a bacterial infection based on machine learning models (MLM). For this purpose, MLM were trained on thousands of measured interactions between a panel of phage and sequenced bacterial isolates. The concept was applied to Escherichia coli ( E. coli ) associated with urinary tract infections. This is an important common infection in humans and companion animals from which multi-drug resistant (MDR) bloodstream infections can originate. The global threat of MDR infection has reinvigorated international efforts into alternatives to antibiotics including phage therapy. E. coli exhibit extensive genome-level variation due to horizontal gene transfer via phage and plasmids. Associated with this, phage selection for E. coli is difficult as individual isolates can exhibit considerable variation in phage susceptibility due to differences in factors important to phage infection including phage receptor profiles and resistance mechanisms. The activity of 31 phage were measured on 314 isolates with growth curves in artificial urine. Random Forest models were built for each phage from bacterial genome features and the more generalist phage, acting on over 20% of the bacterial population, exhibited F1 scores of >0.6 and could be used to predict phage cocktails effective against previously untested strains. The study demonstrates the potential of predictive models which integrate bacterial genomics with phage activity datasets allowing their use on data derived from direct sequencing of clinical samples to inform rapid and effective phage therapy. Significance Statement: With the growing challenge of antimicrobial resistance there is an urgency for alternative treatments for common bacterial diseases including urinary tract infections (UTIs). Escherichia coli is the main causative agent of UTIs in both humans and companion animals with multidrug resistant strains such as the globally disseminated ST131 becoming more common. Bacteriophage (phage) are natural predators of bacteria and potentially an alternative therapy. However, a major barrier for phage therapy is the specificity of phage on target bacteria and therefore difficulty efficiently selecting the appropriate phage. Here, we demonstrate a genomics driven approach using machine learning prediction models combined with phage activity clustering to select phage cocktails based only on the genome sequence of the infecting bacterial strain.


  • Open Access

    Rapid metagenomic sequencing for diagnosis and antimicrobial sensitivity prediction of canine bacterial infections

    by Natalie Ring, Alison S. Low, Bryan Wee, Gavin K. Paterson, Tim Nuttall, David Gally, Richard Mellanby, J. Ross Fitzgerald

    Antimicrobial resistance is a major threat to human and animal health. There is an urgent need to ensure that antimicrobials are used appropriately to limit the emergence and impact of resistance. In the human and veterinary healthcare setting, traditional culture and antimicrobial sensitivity testing typically requires 48–72 h to identify appropriate antibiotics for treatment. In the meantime, broad-spectrum antimicrobials are often used, which may be ineffective or impact non-target commensal bacteria. Here, we present a rapid, culture-free, diagnostics pipeline, involving metagenomic nanopore sequencing directly from clinical urine and skin samples of dogs. We have planned this pipeline to be versatile and easily implementable in a clinical setting, with the potential for future adaptation to different sample types and animals. Using our approach, we can identify the bacterial pathogen present within 5 h, in some cases detecting species which are difficult to culture. For urine samples, we can predict antibiotic sensitivity with up to 95 % accuracy. Skin swabs usually have lower bacterial abundance and higher host DNA, confounding antibiotic sensitivity prediction; an additional host depletion step will likely be required during the processing of these, and other types of samples with high levels of host cell contamination. In summary, our pipeline represents an important step towards the design of individually tailored veterinary treatment plans on the same day as presentation, facilitating the effective use of antibiotics and promoting better antimicrobial stewardship.


  • Open Access

    Rapid metagenomic sequencing for diagnosis and antimicrobial sensitivity prediction of canine bacterial infections

    by Natalie Ring, Alison S. Low, Bryan Wee, Gavin K. Paterson, Tim Nuttall, Richard Mellanby, David Gally, J. Ross Fitzgerald

    Antimicrobial resistance is one of the greatest current threats to human and animal health. There is an urgent need to ensure that antimicrobials are used appropriately to limit the emergence and impact of resistance. In the human and veterinary healthcare setting, traditional culture and antimicrobial sensitivity testing is typically conducted, requiring 48-72 h to identify appropriate antibiotics for treatment. In the meantime, broad-spectrum antimicrobials are often used, which may be ineffective or impact non-target commensal bacteria. Here, we present a rapid diagnostics pipeline, involving metagenomic Nanopore sequencing directly from clinical urine and skin samples of dogs. We have optimised this pipeline to be versatile and easily implementable in a clinical setting, with the potential for future adaptation to different sample types and animals. Using our approach, we can identify the bacterial pathogen present in a sample with 100% sensitivity within 5 hours. For urine samples, we can predict antibiotic sensitivity with up to 95% accuracy. However, skin swabs which exhibited lower bacterial abundance and higher host DNA, were less amenable and an additional host depletion step may be required prior to DNA extraction. In summary, our pipeline represents an important step towards the design of individually tailored veterinary treatment plans on the same day as presentation, facilitating effective use of antibiotics and promoting antimicrobial stewardship. Impact statement: Antimicrobial resistance (AMR) is a major threat to veterinary and human healthcare. It is a one-health problem, as humans and dogs are in close contact, require similar antibiotics, and share bacterial pathogens and AMR genes. Limited treatments options due to AMR would have a catastrophic effect. The risk of infection would render much of modern healthcare (including critical care, orthopaedic and complex surgeries, implants and oncology) impossible. In addition, routine infections could become life threatening. It is therefore critical to preserve the efficacy of these drugs for the future. Inappropriate antimicrobial use is the single biggest factor driving AMR. Antimicrobial stewardship involves reducing antimicrobial use, using first-line narrow-spectrum drugs, and avoiding overly long treatment. Delays in culture-based diagnosis lead clinicians to speculatively use broad-spectrum antibiotics and prolong courses of treatment beyond clinical cure. Our rapid diagnostic approach will have a major impact in reducing, refining and replacing antibiotic use. This will advance antimicrobial stewardship in veterinary and human healthcare. Data summary: All sequencing data mentioned in this work is available from NCBI, BioProject PRJNA925092, Biosamples SAMN32880396 to SAMN32880438, run accessions SRR23195371 to SRR23195413. The authors confirm all supporting data, code and protocols have been provided within the article or through supplementary data files .


  • Open Access

    SARS-CoV-2 RNA levels in Scotland’s wastewater

    by Livia C. T. Scorza, Graeme J. Cameron, Roisin Murray-Williams, David Findlay, Julie Bolland, Brindusa Cerghizan, Kirsty Campbell, David Thomson, Alexander Corbishley, David Gally, et al.

    Nationwide, wastewater-based monitoring was newly established in Scotland to track the levels of SARS-CoV-2 viral RNA shed into the sewage network, during the COVID-19 pandemic. We present a curated, reference data set produced by this national programme, from May 2020 to February 2022. Viral levels were analysed by RT-qPCR assays of the N1 gene, on RNA extracted from wastewater sampled at 122 locations. Locations were sampled up to four times per week, typically once or twice per week, and in response to local needs. We report sampling site locations with geographical coordinates, the total population in the catchment for each site, and the information necessary for data normalisation, such as the incoming wastewater flow values and ammonia concentration, when these were available. The methodology for viral quantification and data analysis is briefly described, with links to detailed protocols online. These wastewater data are contributing to estimates of disease prevalence and the viral reproduction number (R) in Scotland and in the UK.