Biology subjects

Fowler, P. W.

Publications and source records attributed to Fowler, P. W..

At least 19 recordsLinked to original sources

bioRxiv2026

Long-read sequencing of Mycobacterial tuberculosis is comparable to short-read sequencing for antimicrobial resistance prediction and epidemiological studies.

BackgroundShort-read genetic sequencing technologies (mainly Illumina) have been extensively used for around a decade for Mycobacterium tuberculosis complex (MTBC) outbreak analysis and genomic drug susceptibility testing (gDST) with the result that Illumina has become the de facto gold standard. Long-read sequencing, as exemplified by Oxford Nanopore Technologies (ONT), offer the prospect of faster, simpler, and portable sequencing. In this work, we carry out the largest to date comparison of how well Illumina and ONT technologies sequence MTBC samples, making use of R10.4 ONT flowcells, updated basecalling models and deep-learning variant calling. MethodsA total of 508 samples were sequenced using both short and long-read platforms. All samples originated from South Africa or Vietnam and were over-selected for drug resistance and also included several local outbreaks and a range of lineages. The South African and Vietnamese samples had already been Illumina sequenced. Samples with [≥]50 read depth by Illumina were selected for sequencing by ONT using one of the GridION or PromethION platforms. Bioinformatics processing was done using a modified online cloud platform which included reference-based variant calling, catalogue-based gDST and identified related samples via SNP counting to inform outbreak detection. The lineages and gDST predictions obtained by short-and long-sequencing were compared for all samples as were all putative clusters identified via SNP counting. For convenience Illumina was used as the reference method. FindingsOf the 508 samples, 425 (83.7%) had sufficient read depths to permit comparison between the two sequencing technologies. The assigned lineages were identical for 407/425 (95.8%) samples and all discordances were due to mixed lineages being identified by one technology. Evidence of non-tuberculous mycobacterium (NTM) subpopulations were found in nine samples. Using Illumina as the reference method, the very major error (VME) rate of ONT for predicting resistance to all 15 drugs is 1.0% (0.6-1.5%) whilst the major error (ME) rate is 1.7% (1.3-2.2%) with an unclassified rate of 6.9% (6.3-7.5%). This is below the thresholds specified by the CLSI. Considering each of the 15 drugs individually they had VME and ME point estimates below [≤]3% in 29/30 cases; and most 25/30 below [≤]1.5%. Filtering out all samples containing mixtures left 382 isolates. By appropriate masking of the reference genome we were able to obtain a mean SNP distance between the two platforms of 0.13 (median of zero) for the same sample and for 376/382 samples (98.4%, CI:96.6-99.4%) the difference was [≤]1 SNPs. The high concordance in SNP identification ensured that few differences in the 43 putative clusters among 172 isolates were observed. InterpretationThe differences between the two sequencing platforms for the key clinical outputs is so small that it is now within the tolerances set by regulatory agencies. Provided the sequencing is of sufficient quality, we have therefore reached a threshold whereby sequencing data from long-and short-read platforms can be aggregated. This will enable large scale analyses by national and international public health agencies whilst allowing the MTBC community to take advantage of the portability and speed of long-read sequencing. FundingThe NIHR Health Protection Research Unit: Healthcare Associated Infections and Antimicrobial Resistance at University of Oxford (NIHR200915), a partnership between the UK Health Security Agency (UKHSA) and the University of Oxford, the National Institute for Health and Care Research Biomedical Research Centre: Oxford (BRC) and the Ellison Institute of Technology, Oxford Ltd. The CRyPTIC project was funded by Wellcome [214560/Z/18/Z], a Wellcome Trust/Newton Fund-MRC Collaborative Award (200205/Z/15/Z); and the Bill & Melinda Gates Foundation Trust (OPP1133541). Research in contextO_ST_ABSEvidence before this studyC_ST_ABSWe conducted a PubMed Central full text search for "tuberculosis" AND ("drug resistance prediction" OR "drug susceptibility prediction") AND ("genome" OR "genomic" OR "geno-typic") AND ("ont" OR "oxford nanopore") between 2022 and 2026 (conducted 1 April 2026). This returned 62 papers; of which, six used both Illumina and ONT sequencing. One of these, published in 2023, directly compared the performance of the two platforms on 151 M. tuberculosis isolates oversampled for resistance. The investigation yielded comparative results for the earlier generation ONT flow cell (R9{middle dot}4{middle dot}1) and base-caller (guppy version 5{middle dot}0{middle dot}16). Another, published in 2026, investigated a targeted next-generation sequencing panel of 20 amplicons using ONT sequencing on R10.4.1 flow cells with guppy 6{middle dot}4{middle dot}6. They compared the results on 71 isolates against phenotypic data and Illumina whole genome sequencing (for 53 isolates) but had low rates of resistance, with all drugs but isoniazid being limited to under five resistant isolates. Two other small studies (10 and 13 samples, respectively) conducted feasibility studies comparing ONT with Illumina, also using earlier generation flow cells and base-calling technology from ONT. Two further studies compared Illumina with ONT for direct sputum sequencing and did not investigate the comparative performance of the two platforms for variant call accuracy, resistance prediction, and outbreak detection. Illumina sequencing technology is widely used for genomic sequence analysis in research, and clinical and public health contexts. Consequently, it has become the de facto reference standard for generating whole genome sequence data. Whilst previous studies established the promise and limitations of long-read (ONT) sequencing as an alternative to short-read sequencing (mainly Illumina), the enhanced performance arising from newer flowcells (e.g. R10.4.1), V14 chemistry, and the latest basecallers (dorado v4.3.0/5.0.0) has not been analysed. Neither has any ONT analysis incorporating the new deep-learning variant callers been evaluated in a large-scale comparative study. Thus, it is currently unclear whether data generated by either platform can be used safely in aggregated analyses for research and clinical or public health service. Added value of this studyWe compared how well short-(Illumina) and long-read (ONT) sequencing platforms identify the genetic variants in M. tuberculosis, predict antituberculous drug resistance and recog-nise outbreaks. The long-reads were generated using the latest generation ONT R10.4.1 flows cells, V14 chemistry, super high accuracy basecalling (dorado v4.3.0/5.0.0) and a bioinformatics analysis pipeline built using the Clair3 deep-learning based variant caller. A total of 508 clinical samples were sequenced using both technologies, substantially more than previous studies. The sampling frame was much larger than previously investigations and included a large proportion of isolates with resistance to first-line and second-line antibiotics as well as bedaquiline. Thus, providing greater statistical power for resistance prediction than before. In particular, the inclusion of bedaquiline resistance provided evidence useful for predicting resistance to this newly deployed drug for treating multi-drug resistant (MDR) TB. We find that the differences between technologies are small meaning that either technology can be used alone safely, and services using both technologies can confidently aggregate the data for analysis. Implications of all the available evidenceThis will be a benefit to local, regional and international organisations, particularly public health agencies, which often have a mix of the two main sequencing technologies for characterising TB whole genome sequences. It also opens up the sequence based diagnostic market to greater competition, particularly if the observed performance can be replicated for other pathogen species.

microbiology↗

bioRxiv2026

Evaluation of an Oxford Nanopore sequencing workflow for mycobacteria from primary MGIT culture

Illumina sequencing of primary MGIT cultures is an established workflow in several reference mycobacteriology laboratories. Oxford Nanopore Technologies (ONT) provides real-time genetic sequencing yielding long reads which help resolve repetitive genomes and is being explored for in-house implementation within diagnostic laboratories. However, low DNA yields from primary MGIT cultures frequently limit the application of ONT workflows, due to high minimum DNA input requirements for library preparation. We validated a modified ONT workflow combining rapid, semi-automated DNA extraction from MGIT cultures with Rapid PCR-Barcoding for whole-genome amplification, and compared its performance with Illumina sequencing for species identification and Mycobacterium tuberculosis complex (MTBC) single-nucleotide polymorphism (SNP) detection. A platform-agnostic analysis pipeline enabled consistent human read removal, taxonomic assignment, and MTBC genomic characterisation. ONT sequencing data was subsampled at 1, 6, and 72 hours to determine the earliest time point for reliable species identification. The concordance between sequencing platforms on species classification was 98.3% (95.8-99.5%) with all differences arising from potential mixed infections. SNP agreement was high, with a mean of 0.3 and a median of 0 SNP differences between sequencing platforms after masking. These findings demonstrate the feasibility of PCR-amplified ONT sequencing as a reliable alternative for routine genomic characterisation of MGIT cultures. IMPORTANCERapid identification of mycobacterial infections is essential for timely patient care and infection control. Many clinical laboratories currently rely on outsourcing sequencing to external reference centres, which adds time and delays the return of clinically actionable results. These services commonly use Illumina sequencing, which, while highly accurate, involves complex workflows and longer turnaround times. Newer technologies offer the potential to generate results in real time, but their use has been limited by the low amount of DNA available from routine culture samples. In this study, we developed an improved workflow that increases the amount of usable DNA and enables reliable, rapid sequencing directly from these samples. This approach allows multiple samples to be processed together and reduces the time needed to obtain results. Importantly, it could enable clinical laboratories to perform sequencing in-house, reducing reliance on external services and improving turnaround times.

molecular biology↗

bioRxiv2025

Predicting pyrazinamide resistance in Mycobacterium tuberculosis using a graph convolutional network

BackgroundPyrazinamide is an important first-line antibiotic for treating tuberculosis, with resistance primarily driven by mutations in the pncA gene. Traditional machine learning models are able to predict pyrazinamide resistance with some success, but are limited in their ability to incorporate 3-dimensional protein structural information. Graph neural networks offer the potential to integrate protein structure and residue-level features to better predict the impact of mutations on drug resistance. ResultsWe trained a graph convolutional network on PncA variants containing missense mutations and evaluated its ability to classify resistance to pyrazinamide. Each PncA variant was represented as an amino acid-level graph, with edges calculated from 3-dimensional spatial proximity, and node features derived from chemical properties and mutation meta-predictors. We used AlphaFold2 to generate predicted structures of the PncA variants, which we used to create the protein graphs. The predicted structures of resistant PncA variants showed greater deviation from the wild-type structure compared to susceptible variants. Our model achieved an F1 score of 81.6 %, sensitivity of 81.6 % and specificity of 80.4 % on the test set and either matched or exceeded the performance of a published set of traditional machine learning models. We show that both structural graph connectivity and node features contribute significantly to model performance. Furthermore, we employ additional train/test dataset splits to demonstrate the GCNs ability to generalise and predict resistance in samples with mutations in unseen positions and structural regions. ConclusionsOur study demonstrates that graph-based deep learning can leverage protein structure and biochemical features to accurately predict antimicrobial resistance, despite being trained on a small dataset with little variation. We present this as a proof-of-concept for these methods to be applied to resistance phenotype prediction in more genetically diverse pathogens to predict the more complex observed patterns of antimicrobial resistance.

microbiology↗

bioRxiv2025

Rapidly and reproducibly building a comprehensive catalogueof resistance-associated variants for M. tuberculosis

BackgroundCatalogues of genetic variants associated with resistance underpin whole-genome sequencing (WGS)-based predictions of drug susceptibility in Mycobacterium tuberculosis, and are essential for molecular diagnostics and surveillance. The current gold standard catalogues are those released by the WHO but the underlying data are not fully released and they are difficult to interpret. Open and reproducible methods would help address these problems, extending the important work already done. MethodsWe have developed an automated method, catomatic, that uses a binomial test to associate informative isolates with resistance or susceptibility, and built a catalogue (catomatic-1) from the same 39,358 samples used to construct the first edition of the WHO catalogue (WHOv1). We performed a sensitivity analysis to optimise statistical and bioinformatic parameters for each drug, and benchmarked catomatic-1 against WHOv1 using an independent Validation Dataset of 14,380 isolates. FindingsBy using simpler statistics, catomatic-1 algorithmically classified 1,329 genetic variants, ranging from five for linezolid to 440 for pyrazinamide. WHOv1 included generalisable rules added by a panel of experts, increasing its predictive coverage, but at the cost of reproducibility. Despite not including such expert rules, catomatic-1 achieves comparable performance for all drugs, with sensitivities for first-line agents above 88% on the independent Validation Dataset. The automated process allowed us to efficiently explore parameter space; for instance, detecting resistant variants with low read support improved the sensitivity for all drugs. InterpretationPerformant resistance catalogues for M. tuberculosis can be built automatically using transparent and reproducible statistical methods. As more data are collected, catalogue content and performance will evolve, highlighting the need for proper versioning, machine/human readability, and open access. This approach demonstrates resistance catalogues used in surveillance and diagnostics can be rapidly and reproducibily updated. FundingThe National Institute for Health and Care Research (NIHR), Engineering and Physics Sciences Research Council (EPSRC) and ORACLE Corporation. Research in contextO_ST_ABSEvidence before this studyC_ST_ABSWe searched PubMed and preprint servers (bioRxiv, medRxiv), and publicly available mutation catalogues for studies linking Mycobacterium tuberculosis genomic variants with drug resistance using whole-genome or targeted sequencing and phenotypic drug-susceptibility testing (pDST). Search terms combined "Mycobacterium tuberculosis", "genome sequencing", "mutation catalogue", "mutation effects", "drug resistance", and individual drug names, with no language or date restriction. We included studies providing paired, clinical genomic and pDST or MIC data, excluding purely in-silico or case-only reports. This work directly builds on methodologies and data published by five prior studies, and makes primary comparisons with the First (WHOv1) and Second (WHOv2) Editions of the WHO Catalogue of mutations in Mycobacterium tuberculosis. Added value of this studyWe developed catomatic, a transparent, reproducible tool for building catalogues of resistance- and susceptibility-associated genetic variants. Trained on the same samples used to build WHOv1 and benchmarked on an independent Validation Dataset, catomatic achieves comparable sensitivity, specificity, and definitive prediction rates to WHOv1 without expert-rule augmentation and despite using simpler statistics. It optimises parameters per drug, produces machine-readable outputs (CSV/JSON), and demonstrates that adjusting read-support thresholds can improve detection of minor resistance subpopulations. Implications of all the available evidenceCatalogues of resistance-associated variants for M. tuberculosis can be rapidly and transparently constructed. Making catalogues available in human/machine-readable formats with uncertainty estimates will improve uptake of WGS for M. tuberculosis surveillance and diagnostics; using a reproducible process permits diagnostic test manufacturers, researchers, clinical and public health laboratories to select the level of statistical support necessitated by their specific use-case, Policymakers should balance the benefits of expert rules against loss of reproducibility. Future work will expand the size of the datasets used, integrate minimum inhibitory concentration data, and establish consensus workflows for routine, transparent catalogue updates.

microbiology↗

bioRxiv2025

Subpopulations in clinical samples of M. tuberculosis can give rise to rifampicin resistance and shed light on how resistance is acquired

ObjectivesWhole genome sequencing (WGS) has become a key tool for diagnosing Mycobacterium tuberculosis (M. tuberculosis) infections, but discrepancies between genotypic and phenotypic drug susceptibility testing can hinder effective treatment and surveillance. This study investigates the impact of resistant subpopulations and compensatory mutations in WGS-based rifampicin resistance prediction. MethodsBased on a dataset of 35,538 clinical M. tuberculosis samples the sensitivity and specificity of resistance classification were evaluated with and without considering subpopulations and compensatory mutations. ResultsBy lowering the fraction of reads required to identify a resistance-associated variant in a sample from 0.90 to 0.05, the sensitivity increased significantly from 94.3% to 96.4% without a significant impact on specificity. Allowing compensatory mutations to predict resistance further lowered the false negative rate. Finally, we found that samples with resistant subpopulations were less likely to be compensated than homogeneous resistant samples. Further analysis of these samples revealed distinct clusters with differing amounts of within-sample diversity, pointing towards different mechanisms of resistance acquisition, such as within-host evolution and secondary infections. ConclusionsOur results indicate that a substantial fraction of false negative calls in WGS-based rifampicin resistance prediction can be explained by masked resistant subpopulations. The genetic diversity within the heterogenous samples is consistent with at least 28% of the rifampicin resistance arising from secondary infections.

microbiology↗

bioRxiv2025

Deep Learning-based Framework for Mycobacterium tuberculosis Bacterial Growth Detection for Antimicrobial Susceptibility Testing

Tuberculosis (TB) kills more people annually than any other pathogen. Resistance is an ever-increasing global problem, not least because diagnostics remain challenging and access limited. 96-well broth microdilution plates offer one approach to high-throughput phenotypic testing, but they can be challenging to read. Automated Mycobacterial Growth Detection Algorithm (AMyGDA) is a software package that uses image processing techniques to read plates, but struggles with plates that exhibit low growth or images of low quality. We developed a new framework, TMAS (TB Microbial Analysis System), which leverages state-of-the-art deep learning models to detect M. tuberculosis growth in images of 96-well microtiter plates. TMAS is designed to measure Minimum Inhibitory Concentrations (MICs) rapidly and accurately while differentiating between true bacterial growth and artefacts. Using 4,018 plate images from the CRyPTIC (Comprehensive Resistance Prediction for Tuberculosis: An International Consortium) dataset to train models and refine the framework, TMAS achieved an essential agreement of 98.8%. TMAS offers a reliable, automated and complementary evaluation to support expert interpretation, potentially improving accuracy and efficiency in tuberculosis drug susceptibility testing (DST). Author summaryTuberculosis (TB) is one of the worlds leading causes of death from infectious diseases, with drug resistance making treatment increasingly difficult. Accurate and timely drug susceptibility testing (DST) is essential to determine which antibiotics remain effective against a given TB strain. A widely used DST method involves 96-well broth microdilution plates, where bacteria are exposed to different antibiotic concentrations to determine the minimum inhibitory concentration (MIC), the lowest concentration that fully prevents bacterial growth. However, manually interpreting these plates is time-consuming and prone to human error, while existing automated methods often struggle with imaging artifacts such as condensation, shadows, and contamination. In this study, we developed the Tuberculosis Microbial Analysis System (TMAS), a machine learning-based tool designed to automate MIC determination from plate images. Training on advanced deep learning models, TMAS can learn to detect bacterial growth with high precision, distinguishing true growth from artefacts such as shadows, bubbles, sediment, condensation and contamination. When tested on a large dataset, TMAS outperformed existing automated methods in accuracy, reliability and efficiency. By reducing the reliance on manual interpretation, TMAS has the potential to streamline TB diagnostics, improve efficiency in laboratories, and improve access to high-quality DST, particularly in resource-limited settings.

microbiology↗

bioRxiv2025

An improved catalogue for whole-genome sequencing prediction of bedaquiline resistance in M. tuberculosis using a reproduciblealgorithmic approach.

Bedaquiline (BDQ) has only been approved for use for a little over a decade yet is a key drug for treating multi-drug resistant tuberculosis, however rising levels of resistance threaten to reduce its effectiveness. Catalogues of mutations associated with resistance to bedaquiline are key to detecting resistance genetically for either diagnosis or surveillance. At present building catalogues requires considerable expert knowledge, often requires the use of complex grading rules, and is an irreproducible process. We developed an automated method, catomatic, that associates genetic variants with resistance (or susceptibility) using a two-tailed binomial test with a stated background rate and applied it to a dataset of 11,867 Mycobacterium tuberculosis samples with whole genome and bedaquline susceptibility testing data. Using this framework we investigated how to best classify variants and the phenotypic significance of minor alleles. The genes mmpS5 and mmpL5 are not directly associated with bedaquline resistance, and our catalogue of Rv0678, atpE, and pepQ variants attains a cross-validated sensitivity and specificity of 79.4 {+/-} 1.8 % and 98.5 {+/-} 0.3%, respectively, for 94 {+/-} 0.4% of samples. Identifying samples with subpopulations containing Rv0678 variants improves sensitivity, and detection thresholds in bioinformatic pipelines should therefore be lowered. By using a more permissive and deterministic algorithm trained on a sufficient number of resistant samples we have reproducibly constructed an AMR catalogue for BDQ resistance-associated variants that is comprehensive and accurate. Impact StatementBedaquiline has recently received global endorsement for tuberculosis treatment, yet the genetic determinants of antimicrobial resistance remain incompletely understood. Existing gold-standard methods for building mutation catalogs lack public accessibility and reproducibility. We introduce catomatic, a reproducible and publicly available method that employs simpler statistics to increase sensitivity to resistance-associated variants. This approach has enabled investigations into mechanisms of resistance, the significance of genetic subpopulations, and key data attributes that influence the ease of classifying effects and the accuracy of BDQ resistance phenotype prediction in clinical samples. We strongly emphasise the utility in using reproducible statistics and sustainably developed software in genetics-focussed microbiology.

microbiology↗

bioRxiv2024

Enhancement and validation of the antibiotic resistance prediction performance of a cloud-based genetics processing platform for Mycobacteria

Tuberculosis remains a global health problem. Making it easier and quicker to identify which antibiotics an infection is likely to be susceptible to will be a key part of the solution. Whilst whole-genome sequencing offers many advantages, the processing of the genetic reads to produce the relevant public health and clinical information is, surprisingly, often the responsibility of the end user which inhibits uptake. Here we characterise how well a freely-available tool we have developed, gnomonicus, predicts the antibiotic resistance profile of a sample (given its variant call file) using our implementation of the second edition of the WHO catalogue of resistance-associated variants (WHOv2). To facilitate this, we have constructed a Diverse Testset of 2,663 publicly-available M. tuberculosis samples which have both genetic and drug susceptibility testing (DST) data. We have chosen to apply the catalogue such that our tool will return a result of (i) Fail if there are insufficient reads at a genetic locus associated with resistance, (ii) Unknown if a genetic variant in a resistance gene not listed in the catalogue is encountered and (iii) Resistant if three or more short-reads support the presence of a resistance-associated variant. The last step increases the sensitivity for all 15 antibiotics but only reaches significance in a few in our testset. Comparing our results to those of TB-Profiler, an existing tool, highlights the different design choices and demonstrates the performance of both tools on our Diverse Testset is comparable. By only considering high confidence DST results we show that gnomonicus, in combination with our translation of WHOv2, achieves sensitivities and specificities in excess of 95% for both isoniazid and rifampicin. Impact StatementWhole genome sequencing clinical samples taken from patients with tuberculosis is a potentially fast and accurate method for determining to which antibiotics the infection will be susceptible. Two barriers need to be overcome; the first, which is knowing which mutations are associated with resistance (or not) to a range of antibiotics is well on the way to be solved thanks to the efforts of the World Health Organization (WHO) who have published extensive catalogues containing lists of such mutations. The second barrier is that the processing of the raw genetic files remains a largely manual process overseen by bioinformaticians. Here we describe gnomonicus, our open-source AMR prediction tool, and report the performance of our translation of the second edition of the WHO catalogue using a carefully designed publicly-available dataset of 2,663 M. tuberculosis samples. We hope that not only will this tool be useful but also that this dataset will be used by other researchers to facilitate comparisons between pipelines, approaches and tools. Data SummaryThe attendant GitHub repository1 allows gnomonicus to be rerun on all 2,663 samples in the Diverse Testset; it therefore includes instructions, the necessary input files (all the variant call files, the version of the WHOv2 catalogue and a link to the H37Rv GenBank file used in this study) and also the output JSON files. The ENA accession numbers for all 2,663 samples (including a bash script to download them) and their corresponding phenotypic drug susceptibility testing results are included with the intention that people can either reproduce our results, or use the same dataset for other analyses. The JSON files output by TB-Profiler have also been added for comparison. The repository contains a series of Juypter notebooks containing Python3 code that allows the user to discover and parse the output JSON files from either tool and save the results as data tables. Other notebooks allow the user to reproduce all the analysis underlying this work, including reproducing the figures and many of the tables.

microbiology↗

bioRxiv2024

Predicting rifampicin resistance in M. tuberculosis using machine learning informed by protein structural and chemical features.

BackgroundRifampicin remains a key antibiotic in the treatment of tuberculosis. Despite advances in cataloguing resistance-associated variants (RAVs), novel and rare mutations in the relevent gene, rpoB, will be encountered in clinical samples, complicating the task of using genetics to predict whether a sample is resistant or not to rifampicin. We have trained a series of machine learning models with the aim of complementing genetics-based drug susceptibility testing. MethodsWe built a Test+Train dataset comprising 219 susceptible mutations and 46 RAVs. Features derived from the structure of the RNA polymerase or the change in chemistry introduced by the mutation were considered, however, only a few, notably the distance from the rifampicin binding site, were found to be predictive on their own. Due to the paucity of RAVs we used Monte Carlo cross-validation with 50 repeats to train four different machine learning models. ResultsAll four models behaved similarly with sensitivities and specificities in the range 0.84-0.88 and 0.94-0.97 although we preferred the ensemble of Decision Tree models as they are easy to inspect and understand. We showed that measuring distances from molecular dynamics simulations did not improve performance. ConclusionsIt is possible to predict whether a mutation in rpoB confers resistance to rifampicin using a machine learning model trained on a combination of structural, chemical and evolutionary features, however performance is moderate and training is complicated by the lack of data.

microbiology↗

bioRxiv2024

Addressing pandemic-wide systematic errors in the SARS-CoV-2 phylogeny

The SARS-CoV-2 genome occupies a unique place in infection biology - it is the most highly sequenced genome on earth (making up over 20% of public sequencing datasets) with fine scale information on sampling date and geography, and has been subject to unprecedented intense analysis. As a result, these phylogenetic data are an incredibly valuable resource for science and public health. However, the vast majority of the data was sequenced by tiling amplicons across the full genome, with amplicon schemes that changed over the pandemic as mutations in the viral genome interacted with primer binding sites. In combination with the disparate set of genome assembly workflows and lack of consistent quality control (QC) processes, the current genomes have many systematic errors that have evolved with the virus and amplicon schemes. These errors have significant impacts on the phylogeny, and therefore over the last few years, many thousands of hours of researchers time has been spent in "eyeballing" trees, looking for artefacts, and then patching the tree. Given the huge value of this dataset, we therefore set out to reprocess the complete set of public raw sequence data in a rigorous amplicon-aware manner, and build a cleaner phylogeny. Here we provide a global tree of 4,471,579 samples, built from a consistently assembled set of high quality consensus sequences from all available public data as of June 2024, viewable at https://viridian.taxonium.org. Each genome was constructed using a novel assembly tool called Viridian (https://github.com/iqbal-lab-org/viridian), developed specifically to process amplicon sequence data, eliminating artefactual errors and mask the genome at low quality positions. We provide simulation and empirical validation of the methodology, and quantify the improvement in the phylogeny. We hope the tree, consensus sequences and Viridian will be a valuable resource for researchers.

bioinformatics↗

bioRxiv2023

Compensatory mutations are associated with increased in vitro growth in resistant clinical samples of Mycobacterium tuberculosis.

Mutations in Mycobacterium tuberculosis associated with resistance to antibiotics often come with a fitness cost for the bacteria. Resistance to the first-line drug rifampicin leads to lower competitive fitness of M. tuberculosis populations when compared to susceptible populations. This fitness cost, introduced by resistance mutations in the RNA polymerase, can be alleviated by compensatory mutations (CMs) in other regions of the affected protein. CMs are of particular interest clinically since they could lock in resistance mutations, encouraging the spread of resistant strains worldwide. Here, we report the statistical inference of a comprehensive set of CMs in the RNA polymerase of M. tuberculosis, using over 70,000 M. tuberculosis genomes that were collated as part of the CRyPTIC project. The unprecedented size of this data set gave the statistical tests to investigate the association of putative CMs with resistance-conferring mutations much more power. Overall, we propose 51 high-confidence CMs by means of statistical association testing and suggest hypotheses for how they exert their compensatory mechanism by mapping them onto the protein structure. In addition, we were able to show an association of CMs with higher in vitro growth densities, and hence presumably with higher fitness, in resistant samples in the more virulent M. tuberculosis Lineages 2 and 3. In Lineage 2, our results even suggest the association of CM presence with significantly higher in vitro growth than for wild-type samples, although this association is confounded with lineage and sub-lineage affiliation. Our findings emphasise the integral role of CMs and lineage affiliation in resistance spread and increases the urgency for antibiotic stewardship, which implies accurate, cheap and widely accessible diagnostics for M. tuberculosis infections to not only improve patient outcomes but also to prevent the spread of resistant strains.

microbiology↗

bioRxiv2022

Inclusion of minor alleles improves catalogue-based prediction of fluoroquinolone resistance in Mycobacterium tuberculosis

SynopsisFluoroquinolone resistance poses a threat to the successful treatment of tuberculosis. Whole genome sequencing (WGS), and the subsequent detection of catalogued resistance-associated mutations, offers an attractive solution to fluoroquinolone susceptibility testing. However, the bioinformatic pipelines used often mask the recognition of minor alleles which are implicated in fluoroquinolone resistance. Using the Comprehensive Resistance Prediction for Tuberculosis: an International Consortiums (CRyPTIC) dataset of globally diverse WGS Mycobacterium tuberculosis isolates, with matched minimum inhibitory concentrations for two fluoroquinolone drugs, we show that detecting minor alleles increased the sensitivity of WGS for moxifloxacin resistance prediction from 85.4% to 94.0%, without significantly reducing specificity. We also found no correlation between the proportion of an M. tuberculosis population containing a resistance-conferring allele and the magnitude of resistance. Together our results highlight the importance of detecting minor resistance conferring alleles when using WGS, or indeed any sequencing-based approach, to diagnose fluoroquinolone resistance.

microbiology↗

bioRxiv2022

ReadItAndKeep: rapid decontamination of SARS-CoV-2 sequencing reads

SummaryViral sequence data from clinical samples frequently contain human contamination, which must be removed prior to sharing for legal and ethical reasons. To enable host read removal for SARS-CoV-2 sequencing data on low-specification laptops, we developed ReadItAndKeep, a fast lightweight tool for Illumina and nanopore data that only keeps reads matching the SARS-CoV-2 genome. Peak RAM usage is typically below 10MB, and runtime less than one minute. We show that by excluding the polyA tail from the viral reference, ReadItAndKeep prevents bleed-through of human reads, whereas mapping to the human genome lets some reads escape. We believe our test approach (including all possible reads from the human genome, human samples from each of the 26 populations in the 1000 genomes data, and a diverse set of SARS-CoV-2 genomes) will also be useful for others. Availability and implementationReadItAndKeep is implemented in C++, released under the MIT license, and available from https://github.com/GenomePathogenAnalysisService/read-it-and-keep.

bioinformatics↗

bioRxiv2021

Discordance between different bioinformatic methods for identifying resistance genes from short-read genomic data, with a focus on Escherichia coli

2.Several bioinformatics genotyping algorithms are now commonly used to characterise antimicrobial resistance (AMR) gene profiles in whole genome sequencing (WGS) data, with a view to understanding AMR epidemiology and developing resistance prediction workflows using WGS in clinical settings. Accurately evaluating AMR in Enterobacterales, particularly Escherichia coli, is of major importance, because this is a common pathogen. However, robust comparisons of different genotyping approaches on relevant simulated and large real-life WGS datasets are lacking. Here, we used both simulated datasets and a large set of real E. coli WGS data (n=1818 isolates) to systematically investigate genotyping methods in greater detail. Simulated constructs and real sequences were processed using four different bioinformatic programs (ABRicate, ARIBA, KmerResistance, and SRST2, run with the ResFinder database) and their outputs compared. For simulations tests where 3,079 AMR gene variants were inserted into random sequence constructs, KmerResistance was correct for 3,076 (99.9%) simulations, ABRicate for 3,054 (99.2%), ARIBA for 2,783 (90.4%) and SRST2 for 2,108 (68.5%). For simulations tests where two closely related gene variants were inserted into random sequence constructs, KmerResistance the correct alleles in 35,338/46,318 (76.3%) ABRicate identified in 11,842/46,318 (25.6%) of simulations, ARIBA in 1679/46,318 (3.6%), and SRST2 in 2000/46,318 (4.3%). In real data, across all methods, 1392/1818 (76%) isolates had discrepant allele calls for at least one gene. Our evaluations revealed poor performance in scenarios that would be expected to be challenging (e.g. identification of AMR genes at <10x coverage, discriminating between closely related AMR gene sequences), but also identified systematic sequence classification (i.e. naming) errors even in straightforward circumstances, which contributed to 1081/3092 (35%) errors in our most simple simulations and at least 2530/4321 (59%) discrepancies in real data. Further, many of the remaining discrepancies were likely "artefactual" with reporting cut-off differences accounting for at least 1430/4321 (33%) discrepants. Comparing outputs generated by running multiple algorithms on the same dataset can help identify and resolve these artefacts, but ideally new and more robust genotyping algorithms are needed. 3. Impact statementWhole-genome sequencing is widely used for studying the epidemiology of antimicrobial resistance (AMR) genes in bacteria; however, there is some concern that outputs are highly dependent on the bioinformatics methods used. This work evaluates these concerns in detail by comparing four different, commonly used AMR gene typing methods using large simulated and real datasets. The results highlight performance issues for most methods in at least one of several simulated and real-life scenarios. However most discrepancies between methods were due to differential labelling of the same sequences related to the assumptions made regarding the underlying structure of the reference resistance gene database used (i.e. that resistance genes can be easily classified in well-defined groups). This study represents a major advance in quantifying and evaluating the nature of discrepancies between outputs of different AMR typing algorithms, with relevance for historic and future work using these algorithms. Some of the discrepancies can be resolved by choosing methods with fewer assumptions about the reference AMR gene database and manually resolving outputs generated using multiple programs. However, ideally new and better methods are needed.

microbiology↗

bioRxiv2021

BashTheBug: a crowd of volunteers reproducibly and accurately measure the minimum inhibitory concentrations of 13 antitubercular drugs from photographs of 96-well broth microdilution plates.

Tuberculosis is a respiratory disease that is treatable with antibiotics. An increasing prevalence of resistance means that to ensure a good treatment outcome it is desirable to test the susceptibility of each infection to different antibiotics. Conventionally this is done by culturing a clinical sample and then exposing aliquots to a panel of antibiotics, thereby determining the minimum inhibitory concentration (MIC) of each drug. Using 96-well broth micro dilution plates with each well containing a lyophilised pre-determined amount of an antibiotic is a convenient and cost-effective way to measure the MICs of several drugs at once for a clinical sample. Although accurate, this is an expensive and slow process that requires highly-skilled and experienced laboratory scientists. Here we show that, through the BashTheBug project hosted on the Zooniverse citizen science platform, a crowd of volunteers can reproducibly and accurately determine the MICs for 13 drugs and that simply taking the median or mode of 11-17 independent classifications is sufficient. There is therefore a potential role for crowds to support (but not supplant) the role of experts in antibiotic susceptibility testing.

microbiology↗

bioRxiv2021

Deciphering Bedaquiline and Clofazimine Resistance in Tuberculosis: An Evolutionary Medicine Approach

Bedaquiline (BDQ) and clofazimine (CFZ) are core drugs for treatment of multidrug resistant tuberculosis (MDR-TB), however, our understanding of the resistance mechanisms for these drugs is sparse which is hampering rapid molecular diagnostics. To address this, we employed a unique approach using experimental evolution, protein modelling, genome sequencing, and minimum inhibitory concentration data combined with genomes from a global strain collection of over 14,151 Mycobacterium tuberculosis complex isolates and an extensive literature review. Overall, 230 genomic variants causing elevated BDQ and/or CFZ MICs could be discerned, with 201 (87.4%) variants affecting the transcriptional repressor (Rv0678) of an efflux system (mmpS5-mmpL5). Structural modelling of Rv0678 suggests four major mechanisms that confer resistance: impairment of DNA binding, reduction in protein stability, disruption of protein dimerization, and alteration in affinity for its fatty acid ligand. These modelling and experimental techniques will improve personalized medicine in an impending drug resistant era.

microbiology↗

bioRxiv2020

Rapid, qualitative prediction of antimicrobial resistance by alchemical free energy methods

The emergence of antimicrobial resistance (AMR) threatens modern medicine and necessitates more personalised treatment of bacterial infections. Sequencing the whole genome of the pathogen(s) in a clinical sample offers one way to improve clinical microbiology diagnostic services, and has already been adopted for tuberculosis in some countries. A key weakness of a genetics clinical microbiology is it cannot return a result for rare or novel genetic variants and therefore predictive methods are required. Non-synonymous mutations in the S. aureus dfrB gene can be successfully classified as either conferring resistance (or not) by calculating their effect on the binding free energy of the antibiotic, trimethoprim. The underlying approach, alchemical free energy methods, requires large numbers of molecular dynamics simulations to be run. We show that a large number (N=15) of binding free energies calculated from a series of very short (50 ps) molecular dynamics simulations are able to satisfactorily classify all seven mutations in our clinically-derived testset. A result for a single mutation could therefore be returned in less than an hour, thereby demonstrating that this or similar methods are now sufficiently fast and reproducible for clinical use.

microbiology↗

bioRxiv2019

Prediction of pyrazinamide resistance in Mycobacterium tuberculosis using structure-based machine learning approaches.

SynopsisO_ST_ABSBackgroundC_ST_ABSPyrazinamide is one of four first-line antibiotics used to treat tuberculosis, however antibiotic susceptibility testing for pyrazinamide is challenging. Resistance to pyrazinamide is primarily driven by genetic variation in pncA, an enzyme that converts pyrazinamide into its active form. MethodsWe curated a dataset of 664 non-redundant, missense amino acid mutations in pncA with associated high-confidence phenotypes from published studies and then trained three different machine learning models to predict pyrazinamide resistance. All models had access to a range of protein structural-, chemical- and sequence-based features. ResultsThe best model, a gradient-boosted decision tree, achieved a sensitivity of 80.2% and a specificity of 76.9% on the hold-out Test dataset. The clinical performance of the models was then estimated by predicting the binary pyrazinamide resistance phenotype of 4,027 samples harboring 367 unique missense mutations in pncA derived from 24,231 clinical isolates. ConclusionsThis work demonstrates how machine learning can enhance the sensitivity/specificity of pyrazinamide resistance prediction in genetics-based clinical microbiology workflows, highlights novel mutations for future biochemical investigation, and is a proof of concept for using this approach in other drugs.

microbiology↗