SAM-TB: a whole genome sequencing data analysis website for detection of Mycobacterium tuberculosis drug resistance and transmission

Clinical characteristics and genomic epidemiological survey of tuberculosis in Wuzhou, China, 2022

Tuberculosis (TB) is a serious respiratory disease posing significant public health threats, such as the variation in Mycobacterium tuberculosis (M.tb) lineages and their associated drug resistance across regions. In 2022, clinical data and culture-positive TB samples were collected from the Third People's Hospital in Wuzhou, China. M.tb drug resistance and lineage were analyzed using whole-genome sequencing, while logistic regression was applied to identify factors influencing patient outcomes. Among 169 strains analyzed, an overall drug resistance rate of 23.1% was observed. Multidrug-resistant or rifampicin-resistant cases constituted 7.7% of the strains. Most strains belonged to lineage 2 (69.8%), followed by lineage 4 (27.8%). Poor treatment adherence, being aged 65 or older, and retreatment emerged as risk factors for unfavorable outcomes. This pioneering survey provides crucial insights into TB patient characteristics, drug resistance patterns, and lineage distribution in Wuzhou, laying a foundation for future targeted TB control strategies in the region.IMPORTANCEIn 2022, tuberculosis (TB) was the second leading cause of death from a single infectious agent worldwide, posing a serious threat to global health. The epidemiological characteristics of TB vary considerably from country to country, and even from region to region within a single country, due to differences in the economy, medical conditions, education, and other factors. Understanding the current status of TB epidemics in the region is important, with practical implications for local diagnosis, treatment, and control. This genomic epidemiological survey has provided a first insight into the characteristics of TB patients, drug resistance rates, prevalence lineage, and factors associated with unfavorable outcomes in Wuzhou, China.

Advantages of updated WHO mutation catalog combined with existing whole-genome sequencing-based approaches for Mycobacterium tuberculosis resistance prediction

BACKGROUND

The WHO recently released a second edition of the mutation catalog for predicting drug resistance in Mycobacterium tuberculosis (MTB). This study evaluated its effectiveness compared to existing whole-genome sequencing (WGS)-based prediction methods and proposes a novel approach for its optimization.

METHODS

We tested the accuracy of five tools-the WHO catalog, TB Profiler, SAM-TB, GenTB, and MD-CNN-for predicting drug susceptibility on a global dataset of 36,385 MTB isolates with high-quality phenotypic drug susceptibility testing (DST) and WGS data. By integrating the genotypic DST predictions of these five tools in an ensemble machine learning framework, we developed an improved computational model for MTB drug susceptibility prediction. We then validated the ensemble model on 860 MTB isolates with phenotypic and WGS data collected in Shenzhen, China (2013-2019) and Valencia, Spain (2014-2016).

RESULTS

Among the five genotypic DST tools for predicting susceptibility to ten drugs, MD-CNN exhibited the highest overall performance (AUC 92.1%; 95% CI 89.8-94.4%). The WHO catalog demonstrated the highest specificity of 97.3% (95% CI 95.8-98.4%), while TB Profiler had the best sensitivity at 79.5% (95% CI 71.8-86.2%). The ensemble machine learning model (AUC 93.4%; 95% CI 91.4-95.4%) outperformed all of the five individual tools, with a specificity of 95.4% (95% CI 93.0-97.6%) and a sensitivity of 84.1% (95% CI 78.8-88.8%), principally due to considerable improvements in second-line drug resistance predictions (AUC 91.8%; 95% CI 89.6-94.0%).

CONCLUSIONS

The second edition of the WHO MTB mutation catalog does not, by itself, perform better than existing tools for predicting MTB drug resistance. An integrative approach combining the WHO catalog with other genotypic DST methods significantly enhances prediction accuracy.

Cross-District Transmission of Tuberculosis in a High-Mobility City in China: Implications for Regional Collaboration in Infectious Disease Control

Objective

This study aims to elucidate the transmission dynamics of tuberculosis in a Chinese city with high population mobility and to identify the associated risk factors.

Methods

We included the data from ten city-level surveillance sites in Shenzhen between 2018 and 2023. Genomic clusters were defined as having a genomic distance of 12 single nucleotide polymorphisms based on whole-genome sequencing. Cross-district clusters were characterized as clusters containing patients from at least two districts, indicating cross-district transmission. Risk factors for clustering were identified using logistic regression.

Results

Of the 2,519 enrolled patients, 263 (10.4%) were grouped into 119 genomic clusters. Notably, 52.1% (62/119) of these clusters were cross-district clusters. We analyzed the data from Shenzhen's 10 districts separately and compared the results with a citywide combined analysis, finding that the combined analysis revealed significantly higher clustering rates across all districts (P<0.001). Furthermore, the risk of cross-district transmission was 3.41 times higher (95% CI: 1.49-7.80) among internal migrants than among residents. Multivariable logistic regression analysis identified significant risk factors for TB transmission, including age under 25 years (OR=3.07, 95% CI: 1.17-8.03), age 25-44 years (OR=2.86, 95% CI: 1.13-7.23), and drug-resistant TB (OR=1.57, 95% CI: 1.15-2.13).

Conclusion

Cross-district transmission is a key factor in the spread of tuberculosis in cities with high population mobility. TB control institutions at all levels must transcend regional boundaries and enhance collaboration to achieve more effective tuberculosis control.

A duplex one-step recombinase aided PCR assay for the rapid and sensitive detection of the isoniazid resistance genes katG and inhA in Mycobacterium tuberculosis

Objectives

Drug resistance in tuberculosis seriously affects the eradication of tuberculosis, and isoniazid resistance is the second most commonly observed drug resistance in patients with tuberculosis. Timely and accurate detection of isoniazid resistance is critical to the treatment of tuberculosis.

Methods

A duplex one-step recombinase-aided PCR (DO-RAP) assay was developed for the rapid and sensitive detection of the katG Ser315Thr and inhA-15 (C-T) mutations in Mycobacterium tuberculosis, which are the most common isoniazid-resistant mutations. Quantitative recombinant plasmids were used to evaluate the sensitivity of DO-RAP, and 91 Mycobacterium tuberculosis strains with different genotypes, as well as 5 common respiratory tract bacteria, were used to evaluate the specificity of DO-RAP. A total of 78 sputum specimens were simultaneously detected using DO-RAP, quantitative PCR (qPCR) and sanger sequencing of nested PCR products. Sanger sequencing results were used as the standard to verify the clinical performance of DO-RAP.

Results

The reaction time of DO-RAP was less than 1 h. The sensitivity of DO-RAP was 2 copies/reaction, which was 10 times higher than qPCR. The sensitivity of DO-RAP for detecting heterogenous resistance was 5%. There was no cross-reactivity between the isoniazid wild-type gene, drug-resistant mutant genes, and other common respiratory tract bacteria. Compared with Sanger sequencing, the sensitivity, specificity, PPV and NPV of DO-RAP were all 100%. There were 7 specimens with gray zone or negative qPCR results but positive DO-RAP test results.

Conclusion

The DO-RAP can be adopted in ordinary qPCR equipment for the rapid, highly sensitive and specific detection of the isoniazid resistance genes of Mycobacterium tuberculosis.

Machine learning-based approach for identification of new resistance associated mutations from whole genome sequences of Mycobacterium tuberculosis

Motivation

Currently available methods for the prediction of genotypic drug resistance in Mycobacterium tuberculosis utilize information on known markers of drug resistance. Hence, machine learning approaches are needed that can discover new resistance markers.

Results

Whole genome sequences with known phenotypic drug resistance profiles have been utilized to train XGBoost and ANN classifiers for 5 first-line and 8 second-line tuberculosis drugs. Benchmarking on a completely independent dataset from CRyPTIC database revealed that our method has high sensitivity (90%-95%) and specificity (94%-99%) for five first-line drugs and robust performance for six second-line drugs with a sensitivity of 77%-89% at over 95% specificity. An explainable AI method, SHapley Additive exPlanations, has successfully identified resistance mutations for each drug in a completely automated way. This approach could not only identify known resistance associated mutations in agreement with the WHO mutation catalogue, but also predicted >100 other potential resistance associated mutations for 13 antibiotics in new genes outside the known resistance loci. Identification of new resistance markers opens up the opportunity for the discovery of novel mechanisms of drug resistance.

Availability and implementation

Our prediction method has been implemented as TB-AMRpred webserver and command line tool, available freely at http://www.nii.ac.in/TB-AMRpred.html and https://github.com/Ankitapal1995/TB-AMRprd.

Detection of I491F and V170F rpoB mutations associated with misdiagnosis of rifampicin resistance among patients with drug-susceptible tuberculosis treatment failure, Myanmar, 2022

OBJECTIVE

Detecting rifampicin (RIF) resistance is crucial in selecting tuberculosis (TB) treatment. Recently, several studies reported that I491F and V170F rpoB mutations were found with a varying prevalence. This study aimed to find out RIF resistance missed by routine diagnostic assays using next generation genome sequencing tool.

METHODS

Sputum specimens from first-line TB treatment failed patients attending Tuberculosis Centers in Yangon Region during 2022 were cultured in solid media. Phenotypic drug susceptibility testing was conducted using Mycobacterial Growth Indicator Tube method. Whole genome or Deeplex-targeted next-generation sequencing was performed using Illumina Miseq. Mutation analysis was done by PhyResSE and SAM-TB online platforms.

RESULTS

A total of 32 culture-positive isolates with DNA qualified for genome sequencing were included in the study. Those were diagnosed as rifampicin-susceptible by routine GeneXpert and line probe assays. RIF resistance-conferring mutations were found in 17/32 (53.1%) Mycobacterium tuberculosis isolates; 14 (43.7%) had mutations outside the RIF resistance determining region (I491F and V170F), two (6.3%) were S450L, mutation within RIF resistance determining region, and one isolate (3.1%) with interim resistance mutations S428T and S441A.

CONCLUSION

This study highlighted the presence of rifampicin-resistant TB strains missed by current diagnostic strategies, and are circulating as treatment-failed patients. This demonstrates a gap in current World Health Organization-endorsed algorithms for capturing all multidrug-resistant-TB strains.

Global epidemiology of Mycobacterium tuberculosis lineage 4 insights from Ecuadorian genomic data

Tuberculosis is a global public health concern, and understanding Mycobacterium tuberculosis transmission routes and genetic diversity of M. tuberculosis is crucial for outbreak control. This study aimed to explore the genomic epidemiology and genetic diversity of M. tuberculosis in Ecuador by analyzing 88 local isolates and 415 public genomes from 19 countries within the Euro-American lineage (L4). Our results revealed significant genomic diversity among the isolates, particularly in the genes related to protein processing, carbohydrate metabolism, lipid metabolism, and xenobiotic biodegradation and metabolism. The population structure analysis showed that sub-lineages 4.3.2/3 (35.4%), 4.1.2.1 (22.7%), 4.4.1 (12.7%), and 4.1.1. (10.7%) were the most prevalent. Phylogenetic and transmission network analyses suggest that these isolates circulating within Ecuador share genetic ties with isolates from other continents, implying historical and ongoing intercontinental transmission events. Our findings underscore the importance of integrating genomic data into public health strategies for tuberculosis control and suggest that enhanced genomic surveillance is essential for understanding and mitigating the global spread of M. tuberculosis. This study provides a comprehensive genomic framework for future epidemiological investigations and control measures targeting M. tuberculosis L4 in Ecuador.

Decoding Ecuadorian Mycobacterium tuberculosis Isolates: Unveiling Lineage-Associated Signatures in Beta-Lactamase Resistance via Pangenome Analysis

Background: Tuberculosis is the second largest public health threat caused by pathogens. Understanding Mycobacterium tuberculosis's transmission, virulence, and resistance profile is crucial for outbreak control. This study aimed to investigate the pangenome composition of Mycobacterium tuberculosis clinical isolates classified as L4 derived from Ecuador. Methods: We analyzed 88 clinical isolates of Mycobacterium tuberculosis by whole-genome sequencing (WGS) and bioinformatic tools for Lineage, Drug-resistance and Pangenome analysis. Results: In our analysis, we identified the dominance of the LAM lineage (44.3%). The pangenomic analysis revealed a core genome of approximately 3200 genes and a pangenome that differed in accessory and unique genes. According to the COG database, metabolism-related genes were the most representative of all partitions. However, differences were found within all lineages analyzed in the metabolic pathways described by KEGG. Isolates from Ecuador showed variations in genomic regions associated with beta-lactamase susceptibility, potentially leading to epistatic resistance to other drugs commonly used in TB treatment, warranting further investigation. Conclusions: Our findings provide valuable insights into the genetic diversity of Mycobacterium tuberculosis populations in Ecuador. These insights may be associated with increasing adaptation within host heterogeneity, variable latency periods, and reduced host damage, collectively contributing to disease spread. The application of WGS is essential to elucidating the epidemiology of TB in the country.

Molecular typing of Mycobacterium tuberculosis: a review of current methods, databases, softwares, and analytical tools

Studies on the epidemiology and clinical relevance of Mycobacterium tuberculosis complex (MTBC) have immensely benefited from molecular typing methods, associated software applications, and bioinformatics tools. Over the last two decades, the Pasteur Institute of Guadeloupe has developed a range of bioinformatic resources, including databases and software, to advance understanding of TB epidemiology. Traditional methods, such as IS6110-RFLP, MIRU-VNTR typing, and spoligotyping, have been instrumental but are increasingly supplanted by more precise and high-throughput techniques. These typing methods offer relatively good discrimination and reproducibility, making them popular choices for epidemiological studies. However, the advent of whole-genome sequencing (WGS) has revolutionized Mycobacterium tuberculosis complex (MTBC) typing, providing unparalleled resolution and data analysis depth. WGS enables the identification of single nucleotide polymorphisms and other genetic variations, facilitating robust phylogenetic reconstructions, and detailed outbreak investigations. This review summarizes current molecular typing methods, as well as databases and software tools used for MTBC data analysis. A comprehensive comparison of available tools and databases is provided to guide future research on the epidemiology of TB and pathogen-associated variables (drug resistance or virulence) and public health initiatives.

Internal migrants as drivers of long-distance cross-regional transmission of tuberculosis in China

OBJECTIVES

Internal migrants in China frequently travel between their hometowns and the cities where they work, creating ample opportunities for cross-regional transmission of tuberculosis (TB). The aim of this study was to explore the role of internal migrants in transmitting TB across different regions and the contribution of cross-region transmission to China's TB burden.

METHODS

The study included a total of 8664 patients with TB and their Mycobacterium tuberculosis isolates, collected from two large cities and three rural regions. Genomic clusters were defined as having a genomic distance of ≤12-single nucleotide polymorphisms. Cross-regional clusters were defined as clusters containing patients from at least two regions, indicative of cross-regional transmission.

RESULTS

A total of 2403 clustered cases (27.7%) were grouped into 845 clusters, of which 142 (16.8%) were cross-regional. An increased risk for cross-regional transmission was found for internal migrants (adjusted OR (aOR), 1.45; 95% CI, 1.13-1.87), individuals aged <55 years (aOR, 2.73; 95% CI, 1.81-4.13), and housekeepers/factory workers (aOR, 1.16; 95% CI, 0.90-1.50). Among 200 cross-regional transmission events identified by transmission inference, 96 occurred between urban patients, 98 between urban and rural patients, and only six between rural patients. Notably, 93.5% (187/200) of cross-regional transmission events involved internal migrants. Epidemiological data showed that just 5.5% of cross-regional transmission events involved patients from the same township or neighbouring counties, where the transmission likely occurred.

DISCUSSION

The mobility of the internal migrant population appears to be responsible for most cross-regional transmission of TB in China. The magnitude and dynamics of cross-regional transmission should be addressed in future strategies to reduce the incidence of TB in China.

Transmission of tuberculosis in rural Henan, China: a prospective population-based genomic spatial epidemiological study

The incidence of tuberculosis (TB) has declined more slowly in rural than urban areas in China, and data on the patterns of transmission and the high-risk populations in rural areas remains scarce. We conducted a population-based study of culture-positive pulmonary TB patients diagnosed in rural Linzhou City, Henan Province from July 2018 to February 2023. Genomic clusters were defined based on whole-genome sequencing and risk factors for clustering were identified by logistic regression. Transmission events were inferred with phybreak and transmission links were sought through epidemiological investigation of clustered patients. Logistic regression was used to explore the relationship between genomic differences of patient isolates and geographical distances of patient residences. Spatial hotspots were defined using kernel density estimation. Of 455 culture-positive patients, 430 were included in the final analysis. Overall, 192 (44.7%,192/430) patients were grouped into 49 clusters. Clusters containing ≥5 patients accounted for 18.4% (9/49) of the clusters and clustering was highest in student patients. No super-spreaders were detected. Confirmed epidemiologic links were identified for only 18.2% of clustered patients. The clustering risk decreased rapidly with increasing distances between patient residences, but 77.6% of clustered patient pairs lived ≥5.0 km apart. Both the Central Subdistrict and Rencun Township were identified as hotspots for TB transmission. Recent transmission appears to be an important driver of the TB burden in Linzhou. The formulation of effective strategies to reduce TB incidence in rural areas will require further studies to identify high-risk populations and venues where local inhabitants congregate and transmit the infection.

Implications for TB control among migrants in large cities in China: a prospective population-based genomic epidemiology study in Shenzhen

Internal migrants are a challenge for TB control in large Chinese cities and understanding this epidemiology is crucial for designing effective control and prevention strategies. We conducted a prospective genomic epidemiological study of culture-positive TB patients diagnosed between June 1, 2018 and May 31, 2021 in the Longhua District of Shenzhen. Treatment status was obtained from local and national TB registries and all isolates were sequenced. Genomic clusters were defined as strains differing by ≤12 SNPs. Risk factors for clustering were identified with multivariable analysis and then Bayesian models and TransPhylo were used to infer the timing of transmission within clusters. Of the 2277 culture-positive patients, 70.1% (1596/2277) were migrants: 72.1% (1043/1446) of the migrants patients developed TB within two years of arriving in Longhua; 38.8% within 6 months of arriving; and 12.3% (104/843) had TB symptoms when they arrived. Only 15.4% of Longhua strains were in genomic clusters. More than one third (33.6%) of patients were not treated in Shenzhen but were involved in nearly one third of the recent transmission events. Clustering was associated with migrants not treated in Shenzhen, males, and teachers/trainers. TB in Longhua is prinicipally due to reactivation of infections in migrants, but a proportion may have had clinical or incipient TB upon arrival in the district. Patients diagnosed but not treated in Longhua were involved in recent local TB transmission. Controlling TB in Shenzhen will require strategies to comprehensively diagnose and treat active TB in the internal migrant population.

Evaluation of WHO catalog of mutations and five WGS analysis tools for drug resistance prediction of Mycobacterium tuberculosis isolates from China

The continuous advancement of molecular diagnostic techniques, particularly whole-genome sequencing (WGS), has greatly facilitated the early diagnosis of drug-resistant tuberculosis patients. Nonetheless, the interpretation of results from various types of mutations in drug-resistant-associated genes has become the primary challenge in the field of molecular drug-resistance diagnostics. In this study, our primary objective is to evaluate the diagnosis accuracy of the World Health Organization (WHO) catalog of mutations and five WGS analysis tools (PhyResSE, Mykrobe, TB Profiler, Gen-TB, and SAM-TB) in drug resistance to 10 anti-Mycobacterium tuberculosis (MTB) drugs. We utilized the data of WGS collected between 2014 and 2017 in Zhejiang Province, consisting of 110 MTB isolates as detailed in our previous study. Based on phenotypic drug susceptibility testing (DST) results using the proportion method on Löwenstein-Jensen medium with antibiotics, we evaluated the predictive accuracy of genotypic DST obtained by these tools. The results revealed that the WHO catalog of mutations and five WGS analysis tools exhibit robust predictive capabilities concerning resistance to isoniazid, rifampicin, ethambutol, streptomycin, amikacin, kanamycin, and capreomycin. Notably, Mykrobe, SAM-TB, and TB Profiler demonstrate the most accurate predictions for resistance to pyrazinamide, prothionamide, and para-aminosalicylic acid, respectively. These findings are poised to significantly guide and influence future clinical treatment strategies and resistance monitoring protocols.IMPORTANCEWhole-genome sequencing (WGS) has the potential for the early diagnosis of drug-resistant tuberculosis. However, the interpretation of mutations of drug-resistant-associated genes represents a significant challenge as the amount and complexity of WGS data. We evaluated the accuracy of the World Health Organization catalog of mutations and five WGS analysis tools in predicting drug resistance to first-line and second-line anti-TB drugs. Our results offer clinicians guidance on selecting appropriate WGS analysis tools for predicting resistance to specific anti-TB drugs.

An overview of next generation sequencing strategies and genomics tools used for tuberculosis research

Tuberculosis (TB) is a grave public health concern and is considered the foremost contributor to human mortality resulting from infectious disease. Due to the stringent clonality and extremely restricted genomic diversity, conventional methods prove inefficient for in-depth exploration of minor genomic variations and the evolutionary dynamics operating in Mycobacterium tuberculosis (M.tb) populations. Until now, the majority of reviews have primarily focused on delineating the application of whole-genome sequencing (WGS) in predicting antibiotic resistant genes, surveillance of drug resistance strains, and M.tb lineage classifications. Despite the growing use of next generation sequencing (NGS) and WGS analysis in TB research, there are limited studies that provide a comprehensive summary of there role in studying macroevolution, minor genetic variations, assessing mixed TB infections, and tracking transmission networks at an individual level. This highlights the need for systematic effort to fully explore the potential of WGS and its associated tools in advancing our understanding of TB epidemiology and disease transmission. We delve into the recent bioinformatics pipelines and NGS strategies that leverage various genetic features and simultaneous exploration of host-pathogen protein expression profile to decipher the genetic heterogeneity and host-pathogen interaction dynamics of the M.tb infections. This review highlights the potential benefits and limitations of NGS and bioinformatics tools and discusses their role in TB detection and epidemiology. Overall, this review could be a valuable resource for researchers and clinicians interested in NGS-based approaches in TB research.

Strengthening the Diagnosis of Drug-Resistant Tuberculosis Using NGS-Based Approaches and Bioinformatics Pipelines for Data Analysis in India

In India, drug-resistant tuberculosis (DR-TB) is a major public health issue and a significant challenge to stop TB program. An estimated 27% of new TB cases and 44% of previously treated TB cases are resistant to at least one anti-TB drug. The conventional methods for DR-TB diagnosis are time-consuming and have limitations, leading to delays in treatment initiation and the spread of the disease. Next-generation sequencing (NGS) based approaches have emerged as a promising tool for diagnosing DR-TB, simultaneously offering rapid and accurate detection of resistance mutations in multiple genes. NGS-based approaches generate a large amount of data, which requires efficient and reliable bioinformatics pipelines for data analysis. TBProfiler and Mykrobe are the bioinformatics pipelines that have been created to analyze NGS data for the diagnosis of DR-TB. These pipelines use reference-based and machine-learning approaches to detect resistance mutations and predict drug susceptibility, enabling clinicians to make informed treatment decisions. Implementing NGS-based approaches and bioinformatics pipelines for DR-TB diagnosis can potentially improve patient outcomes by facilitating early detection of drug resistance and guiding personalized treatment regimens. However, the widespread adoption of these approaches in India faces several challenges, including high costs, limited infrastructure, and a lack of trained personnel. Addressing these challenges requires concerted effort to ensure equitable access to and effective implementation of these innovative technologies.

Whole-Genome Sequencing Predicting Phenotypic Antitubercular Drug Resistance: Meta-analysis

BACKGROUND

For simultaneous prediction of phenotypic drug susceptibility test (pDST) for multiple antituberculosis drugs, the whole genome sequencing (WGS) data can be analyzed using either a catalog-based approach, wherein 1 causative mutation suggests resistance, (eg, World Health Organization catalog) or noncatalog-based approach using complicated algorithm (eg, TB-profiler, machine learning). The aim was to estimate the predictive ability of WGS-based tests with pDST as the reference, and to compare the 2 approaches.

METHODS

Following a systematic literature search, the diagnostic test accuracies for 14 drugs were pooled using a random-effect bivariate model.

RESULTS

Of 779 articles, 44 with 16 821 specimens for meta-analysis and 13 not for meta-analysis were included. The areas under summary receiver operating characteristic curve suggested test accuracy was excellent (0.97-1.00) for 2 drugs (isoniazid 0.975, rifampicin 0.975), very good (0.93-0.97) for 8 drugs (pyrazinamide 0.946, streptomycin 0.952, amikacin 0.968, kanamycin 0.963, capreomycin 0.965, para-aminosalicylic acid 0.959, levofloxacin 0.960, ofloxacin 0.958), and good (0.75-0.93) for 4 drugs (ethambutol 0.926, moxifloxacin 0.896, ethionamide 0.878, prothionamide 0.908). The noncatalog-based and catalog-based approaches had similar ability for all drugs.

CONCLUSIONS

WGS accurately identifies isoniazid and rifampicin resistance. For most drugs, positive WGS results reliably predict pDST positive. The 2 approaches had similar ability.

CLINICAL TRIALS REGISTRATION

UMIN-ID UMIN000049276.

GenoMycAnalyzer: a web-based tool for species and drug resistance prediction for Mycobacterium genomes

BACKGROUND

Drug-resistant tuberculosis (TB) is a major threat to global public health. Whole-genome sequencing (WGS) is a useful tool for species identification and drug resistance prediction, and many clinical laboratories are transitioning to WGS as a routine diagnostic tool. However, user-friendly and high-confidence automated bioinformatics tools are needed to rapidly identify M. tuberculosis complex (MTBC) and non-tuberculous mycobacteria (NTM), detect drug resistance, and further guide treatment options.

RESULTS

We developed GenoMycAnalyzer, a web-based software that integrates functions for identifying MTBC and NTM species, lineage and spoligotype prediction, variant calling, annotation, drug-resistance determination, and data visualization. The accuracy of GenoMycAnalyzer for genotypic drug susceptibility testing (gDST) was evaluated using 5,473 MTBC isolates that underwent phenotypic DST (pDST). The GenoMycAnalyzer database was built to predict the gDST for 15 antituberculosis drugs using the World Health Organization mutational catalogue. Compared to pDST, the sensitivity of drug susceptibilities by the GenoMycAnalyzer for first-line drugs ranged from 95.9% for rifampicin (95% CI 94.8-96.7%) to 79.6% for pyrazinamide (95% CI 76.9-82.2%), whereas those for second-line drugs ranged from 98.2% for levofloxacin (95% CI 90.1-100.0%) to 74.9% for capreomycin (95% CI 69.3-80.0%). Notably, the integration of large deletions of the four resistance-conferring genes increased gDST sensitivity. The specificity of drug susceptibilities by the GenoMycAnalyzer ranged from 98.7% for amikacin (95% CI 97.8-99.3%) to 79.5% for ethionamide (95% CI 76.4-82.3%). The incorporated Kraken2 software identified 1,284 mycobacterial species with an accuracy of 98.8%. GenoMycAnalyzer also perfectly predicted lineages for 1,935 MTBC and spoligotypes for 54 MTBC.

CONCLUSIONS

GenoMycAnalyzer offers both web-based and graphical user interfaces, which can help biologists with limited access to high-performance computing systems or limited bioinformatics skills. By streamlining the interpretation of WGS data, the GenoMycAnalyzer has the potential to significantly impact TB management and contribute to global efforts to combat this infectious disease. GenoMycAnalyzer is available at http://www.mycochase.org .

Emerging insight of whole genome sequencing coupled with protein structure prediction into the pyrazinamide-resistance signature of Mycobacterium tuberculosis

Pyrazinamide (PZA) is considered to be a pivotal drug to shorten the treatment of both drug-susceptible and drug-resistant tuberculosis, but its use is challenged by the reliability of drug-susceptibility testing (DST). PZA resistance in Mycobacterium tuberculosis (MTB) is relevant to the amino acid substitution of pyrazinamidase that is responsible for the conversion of PZA to active pyrazinoic acid (POA). The single nucleotide variants (SNVs) within ribosomal protein S1 (rpsA) or aspartate decarboxylase (panD), the binding targets of POA, has been reported to drive the PZA-resistance signature of MTB. In this study, whole genome sequencing (WGS) was used to identify SNVs within the pncA, rpsA and panD genes in 100 clinical MTB isolates associated with DST results for PZA. The potential influence of high-confidence, interim-confidence or emerging variants on the interplay between target genes and PZA or POA was simulated computationally, and predicted with a protein structure modelling approach. The DST results showed weak agreement with the identification of high-confidence variants within the pncA gene (Cohen's kappa coefficient=0.58), the analytic results of WGS coupled with protein structure modelling on pncA mutants (Cohen's kappa coefficient=0.524) or related genes (Cohen's kappa coefficient=0.504). Taken together, these results suggest the practicable application of a genotypic-coupled bioinformatic approach to manage PZA-containing regimens for patients with MTB.

TB-DROP: deep learning-based drug resistance prediction of Mycobacterium tuberculosis utilizing whole genome mutations

The most widely practiced strategy for constructing the deep learning (DL) prediction model for drug resistance of Mycobacterium tuberculosis (MTB) involves the adoption of ready-made and state-of-the-art architectures usually proposed for non-biological problems. However, the ultimate goal is to construct a customized model for predicting the drug resistance of MTB and eventually for the biological phenotypes based on genotypes. Here, we constructed a DL training framework to standardize and modularize each step during the training process using the latest tensorflow 2 API. A systematic and comprehensive evaluation of each module in the three currently representative models, including Convolutional Neural Network, Denoising Autoencoder, and Wide & Deep, which were adopted by CNNGWP, DeepAMR, and WDNN, respectively, was performed in this framework regarding module contributions in order to assemble a novel model with proper dedicated modules. Based on the whole-genome level mutations, a de novo learning method was developed to overcome the intrinsic limitations of previous models that rely on known drug resistance-associated loci. A customized DL model with the multilayer perceptron architecture was constructed and achieved a competitive performance (the mean sensitivity and specificity were 0.90 and 0.87, respectively) compared to previous ones. The new model developed was applied in an end-to-end user-friendly graphical tool named TB-DROP (TuBerculosis Drug Resistance Optimal Prediction: https://github.com/nottwy/TB-DROP ), in which users only provide sequencing data and TB-DROP will complete analysis within several minutes for one sample. Our study contributes to both a new strategy of model construction and clinical application of deep learning-based drug-resistance prediction methods.

A precision overview of genomic resistance screening in Ecuadorian isolates of Mycobacterium tuberculosis using web-based bioinformatics tools

INTRODUCTION

Tuberculosis (TB) is among the deadliest diseases worldwide, and its impact is mainly due to the continuous emergence of resistant isolates during treatment due to the laborious process of resistance diagnosis, nonadherence to treatment and circulation of previously resistant isolates of Mycobacterium tuberculosis. In this study, we evaluated the performance and functionalities of web-based tools, including Mykrobe, TB-profiler, PhyResSE, KvarQ, and SAM-TB, for detecting resistance in 88 Ecuadorian isolates of Mycobacterium tuberculosis drug susceptibility tested previously. Statistical analysis was used to determine the correlation between genomic and phenotypic analysis. Our results showed that with the exception of KvarQ, all tools had the highest correlation with the conventional drug susceptibility test (DST) for global resistance detection (98% agreement and 0.941 Cohen's kappa), while SAM-TB, PhyResSE, TB-profiler and Mykrobe had better correlations with DST for first-line drug analysis individually. We also identified that in our study, only 50% of mutations characterized by the web-based tools in the rpoB, katG, embB, pncA, gyrA and rrs regions were canonical and included in the World Health Organization (WHO) catalogue. Our findings suggest that SAM-TB, PhyResSE, TB-profiler and Mykrobe were efficient in determining canonical resistance-related mutations, but more analysis is needed to improve second-line detection. Improving surveillance programs using whole-genome sequencing tools for first-line drugs, MDR-TB and XDR-TB is essential to understand the molecular epidemiology of TB in Ecuador.

IMPORTANCE

Tuberculosis, an infectious disease caused by Mycobacterium tuberculosis, most commonly affects the lungs and is often spread through the air when infected people cough, sneeze, or spit. However, despite the existence of effective drug treatment, patient adherence, long duration of treatment, and late diagnosis have reduced the effectiveness of therapy and increased drug resistance. The increase in resistant cases, added to the impact of the COVID-19 pandemic, has highlighted the importance of implementing efficient and timely diagnostic methodologies worldwide. The significance of our research is in evaluating and identifying a more efficient and user-friendly web-based tool to characterize resistance in Mycobacterium tuberculosis by whole-genome sequencing, which will allow more routine application to improve TB strain surveillance programs locally.

Genotypic and phenotypic comparison of drug resistance profiles of clinical multidrug-resistant Mycobacterium tuberculosis isolates using whole genome sequencing in Latvia

BACKGROUND

Multidrug-resistant tuberculosis (MDR-TB) remains a major public health problem in many high tuberculosis (TB) burden countries. Phenotypic drug susceptibility testing (DST) take several weeks or months to result, but line probe assays and Xpert/Rif Ultra assay detect a limited number of resistance conferring gene mutations. Whole genome sequencing (WGS) is an advanced molecular testing method which theoretically can predict the resistance of M. tuberculosis (Mtb) isolates to all anti-TB agents through a single analysis.

METHODS

Here, we aimed to identify the level of concordance between the phenotypic and WGS-based genotypic drug susceptibility (DS) patterns of MDR-TB isolates. Overall, data for 12 anti-TB medications were analyzed.

RESULTS

In total, 63 MDR-TB Mtb isolates were included in the analysis, representing 27.4% of the total number of MDR-TB cases in Latvia in 2012-2014. Among them, five different sublineages were detected, and 2.2.1 (Beijing group) and 4.3.3 (Latin American-Mediterranean group) were the most abundant. There were 100% agreement between phenotypic and genotypic DS pattern for isoniazid, rifampicin, and linezolid. High concordance rate (> 90%) between phenotypic and genotypic DST results was detected for ofloxacin (93.7%), pyrazinamide (93.7%) and streptomycin (95.4%). Phenotypic and genotypic DS patterns were poorly correlated for ethionamide (agreement 56.4%), ethambutol (85.7%), amikacin (82.5%), capreomycin (81.0%), kanamycin (85.4%), and moxifloxacin (77.8%). For capreomycin, resistance conferring mutations were not identified in several phenotypically resistant isolates, and, in contrary, for ethionamide, ethambutol, amikacin, kanamycin, and moxifloxacin the resistance-related mutations were identified in several phenotypically sensitive isolates.

CONCLUSIONS

WGS is a valuable tool for rapid genotypic DST for all anti-TB agents. For isoniazid and rifampicin phenotypic DST potentially can be replaced by genotypic DST based on 100% agreement between the tests. However, discrepant results for other anti-TB agents limit their prescription based solely on WGS data. For clinical decision, at the current level of knowledge, there is a need for combination of genotypic DST with modern, validated phenotypic DST methodologies for those medications which did not showed 100% agreement between the methods.

Comparison of Xpert MTB/RIF Ultra with Xpert MTB/RIF for the detection of Mycobacterium tuberculosis and rifampicin resistance in a primary-level clinic in rural China

The Xpert MTB/RIF Ultra (Ultra) is not yet used for the diagnosis of tuberculosis (TB) in China. We compared the performance of the Xpert and Ultra for detecting Mycobacterium tuberculosis and rifampicin resistance in a primary-level clinic in rural China. Sputum samples from suspected pulmonary TB patients were collected and subjected to smear microscopy, liquid culture, Xpert and Ultra tests. We then compared the sensitivity and specificity of Xpert and Ultra for diagnosing TB against liquid culture. Whole-genome sequencing was performed to predict rifampicin resistance and the results were compared with the Xpert and Ultra tests. The sensitivities of Xpert and Ultra were 88.1% and 95.1%, and the specificities were 91.9% and 84.4%, respectively. Among the 61 smear-negative culture-positive patients, the sensitivities of Xpert and Ultra were 80.3% and 91.8%. All Xpert-positive patients were Ultra-positive. Among culture-negative Xpert or Ultra-positive patients, 69.6% were taking anti-TB drugs or had a previous history of TB. Of the samples that Ultra classified as trace, nearly 25% were probably false-positives. Both Xpert and Ultra accurately detected all rifampicin-resistant patients. In conclusion, Ultra was more sensitive than Xpert, especially for smear-negative patients but had decreased specificity with more false-positives, especially with Ultra trace results.

Genotypic and spatial analysis of transmission dynamics of tuberculosis in Shanghai, China: a 10-year prospective population-based surveillance study

Background

With improved tuberculosis (TB) control programs, the incidence of TB in China declined dramatically over the past few decades, but recently the rate of decrease has slowed, especially in large cities such as Shanghai. To help formulate strategies to further reduce TB incidence, we performed a 10-year study in Songjiang, a district of Shanghai, to delineate the characteristics, transmission patterns, and dynamic changes of the local TB burden.

Methods

We conducted a population-based study of culture-positive pulmonary TB patients diagnosed in Songjiang during 2011-2020. Genomic clusters were defined with a threshold distance of 12-single-nucleotide-polymorphisms based on whole-genome sequencing, and risk factors for clustering were identified by logistic regression. Transmission inference was performed using phybreak. The distances between the residences of patients were compared to the genomic distances of their isolates. Spatial patient hotspots were defined with kernel density estimation.

Findings

Of 2212 enrolled patients, 74.7% (1652/2212) were internal migrants. The clustering rate (25.2%, 558/2212) and spatial concentrations of clustered and unclustered patients were unchanged over the study period. Migrants had significantly higher TB rates but less clustering than residents. Clustering was highest in male migrants, younger patients and both residents and migrants employed in physical labor. Only 22.1% of transmission events occurred between residents and migrants, with residents more likely to transmit to migrants. The clustering risk decreased rapidly with increasing distances between patient residences, but more than half of clustered patient pairs lived ≥5 km apart. Epidemiologic links were identified for only 15.6% of clustered patients, mostly in close contacts.

Interpretation

Although some of the TB in Songjiang's migrant population is caused by strains brought by infected migrants, local, recent transmission is an important driver of the TB burden. These results suggest that further reductions in TB incidence require novel strategies to detect TB early and interrupt urban transmission.

Funding

Shanghai Municipal Science and Technology Major Project (ZD2021CY001), National Natural Science Foundation of China (82272376), National Research Council of Science and Technology Major Project of China (2017ZX10201302-006).

Whole-Genome Sequencing Exhibits Better Diagnostic Performance than Variable-Number Tandem Repeats for Identifying Mixed Infections of Mycobacterium tuberculosis

Mixed infections of Mycobacterium tuberculosis, defined as the coexistence of multiple genetically distinct strains within a single host, have been associated with unfavorable treatment outcomes. Different methods have been used to detect mixed infections, but their performances have not been carefully evaluated. To compare the sensitivity of whole-genome sequencing (WGS) and variable-number tandem repeats (VNTR) typing to detect mixed infections, we prepared 10 artificial samples composed of DNA mixtures from two strains in different proportions and retrospectively collected 1,084 clinical isolates. The limit of detection (LOD) for the presence of a minor strain was 5% for both WGS and VNTR typing. The overall clinical detection rate of mixed infections was 3.7% (40/1,084) for the two methods combined, WGS identified 37/1,084 (3.4%), and VNTR typing identified 14/1,084 (1.3%), including 11 also identified by WGS. Multivariate analysis demonstrated that retreatment patients had a 2.7 times (95% confidence interval [CI], 1.2 to 6.0) higher risk of mixed infections than new cases. Collectively, WGS is a more reliable tool to identify mixed infections than VNTR typing, and mixed infections are more common in retreated patients. IMPORTANCE Mixed infections of M. tuberculosis have the potential to render treatment regimens ineffective and affect the transmission dynamics of the disease. VNTR typing, currently the most widely used method for the detection of mixed infections, detects mixed infections only by interrogating a small fraction of the M. tuberculosis genome, which necessarily limits sensitivity. With the introduction of WGS, it became possible to study the entire genome, but no quantitative comparison has yet been undertaken. Our systematic comparison of the ability of WGS and VNTR typing to detect mixed infections, using both artificial samples and clinical isolates, revealed the superior performance of WGS at a high sequencing depth (~100×) and found that mixed infections are more common in patients being retreated for tuberculosis (TB) in the populations studied. This provides valuable information for the application of WGS in the detection of mixed infections and the implications of mixed infections for tuberculosis control.

Environment pollutants exposure affects the endogenous activation of within-host Mycobacterium tuberculosis

OBJECTIVE

Epidemiological studies have linked ambient pollutants with tuberculosis (TB) risk, but the association has not been fully understood. Here, for the first time, we applied whole-genome sequencing (WGS) to assess the reproductive state of Mycobacterium tuberculosis (MTB) by profiling the mutation rate of MTB (MTBMR) during within-host endogenous reactivated progression, intending to dissect the actual effects of ambient pollutants on the endogenous reactivation.

METHODS

We conducted a retrospective cohort study on bacteriologically confirmed TB patients and followed them for relapse in Jiangsu and Sichuan Province, China. Endogenous and exogenous activation were distinguished by WGS of the pathogen. The average concentration of air pollution was estimated by considering a lag of 0-1 to 0-12 months. We applied a generalized additive model with a Poisson function to evaluate the relationships between ambient pollutants exposure and MTBMR.

RESULTS

In the single-pollutant adjusted models, the maximum effect for PM₁₀ (MTBMR increase: 81.87%, 95% CI: 38.38, 139.03) and PM_2.5 (MTBMR increase: 73.91%, 95% CI: 22.17, 147.55) was observed at a lag of 0-12 months for every 10 μg/m³ increase. For SO₂, the maximum effect was observed at lag 0-8 months, with MTBMR increasing by 128.06% (95% CI: 45.92, 256.44); and for NO₂, the maximum effect was observed at lag 0-9 months, with MTBMR increasing by 124.02% (95% CI: 34.5, 273.14). In contrast, the O₃ concentration was inversely associated with MTBMR, and the maximum reduction of MTBMR was 6.18% (95% CI: -9.24, -3.02) at a lag of 0-9 months. Similar results were observed for multi-pollutant models.

CONCLUSIONS

Increased exposure to ambient pollutants (PM₁₀, PM_2.5, SO₂, and NO₂) contributed to a faster MTBMR, indicating that MTB exhibits increased reproductive activity, thus accelerating within-host endogenous reactivation. O₃ exposure could decrease the MTBMR, suggesting that MTB exerts low reproductive activity by inhibiting within-host endogenous activation.

Transmission of Mycobacterium tuberculosis in schools: a molecular epidemiological study using whole-genome sequencing in Guangzhou, China

Background

China is a country with a high burden of tuberculosis (TB). TB outbreaks are frequent in schools. Thus, understanding the transmission patterns is crucial for controlling TB.

Method

In this genomic epidemiological study, the conventional epidemiological survey data combined with whole-genome sequencing was used to assess the genotypic distribution and transmission characteristics of Mycobacterium tuberculosis strains isolated from patients with TB attending schools during 2015 to 2019 in Guangzhou, China.

Result

The TB incidence was mainly concentrated in regular secondary schools and technical and vocational schools. The incidence of drug resistance among the students was 16.30% (22/135). The phylogenetic tree showed that 79.26% (107/135) and 20.74% (28/135) of the strains belonged to lineage 2 (Beijing genotype) and lineage 4 (Euro-American genotype), respectively. Among the 135 isolates, five clusters with genomic distance within 12 single nucleotide polymorphisms were identified; these clusters included 10 strains, accounting for an overall clustering rate of 7.4% (10/135), which showed a much lower transmission index. The distance between the home or school address and the interval time of symptom onset or diagnosis indicated that campus dissemination and community dissemination may be existed both, and community dissemination is the main.

Conclusion and recommendation

TB cases in Guangzhou schools were mainly disseminated and predominantly originated from community transmission. Accordingly, surveillance needs to be strengthened to stop the spread of TB in schools.

Bioinformatics toolbox for exploring target mutation-induced drug resistance

Drug resistance is increasingly among the main issues affecting human health and threatening agriculture and food security. In particular, developing approaches to overcome target mutation-induced drug resistance has long been an essential part of biological research. During the past decade, many bioinformatics tools have been developed to explore this type of drug resistance, and they have become popular for elucidating drug resistance mechanisms in a low cost, fast and effective way. However, these resources are scattered and underutilized, and their strengths and limitations have not been systematically analyzed and compared. Here, we systematically surveyed 59 freely available bioinformatics tools for exploring target mutation-induced drug resistance. We analyzed and summarized these resources based on their functionality, data volume, data source, operating principle, performance, etc. And we concisely discussed the strengths, limitations and application examples of these tools. Specifically, we tested some predictive tools and offered some thoughts from the clinician's perspective. Hopefully, this work will provide a useful toolbox for researchers working in the biomedical, pesticide, bioinformatics and pharmaceutical engineering fields, and a good platform for non-specialists to quickly understand drug resistance prediction.

First Insight into Diversity of Minisatellite Loci in Mycobacterium bovis/M. caprae in Bulgaria

The aim of this study was to assess the diversity of minisatellite VNTR loci in Mycobacterium bovis/M. caprae isolates in Bulgaria and view their position within global M. bovis diversity. Forty-three M. bovis/M. caprae isolates from cattle in different farms in Bulgaria were collected in 2015-2021 and typed in 13 VNTR loci. The M. bovis and M. caprae branches were clearly separated on the VNTR phylogenetic tree. The larger and more geographically dispersed M. caprae group was more diverse than M. bovis group was (HGI 0.67 vs. 0.60). Overall, six clusters were identified (from 2 to 19 isolates) and nine orphans (all loci-based HGI 0.79). Locus QUB3232 was the most discriminatory one (HGI 0.64). MIRU4 and MIRU40 were monomorphic, and MIRU26 was almost monomorphic. Four loci (ETRA, ETRB, Mtub21, and MIRU16) discriminated only between M. bovis and M. caprae. The comparison with published VNTR datasets from 11 countries showed both overall heterogeneity between the settings and predominantly local evolution of the clonal complexes. To conclude, six loci may be recommended for primary genotyping of M. bovis/M. caprae isolates in Bulgaria: ETRC, QUB11b, QUB11a, QUB26, QUB3232, and MIRU10 (HGI 0.77). VNTR typing based on a limited number of loci appears to be useful for primary bTB surveillance.

Whole-genome sequencing to characterize the genetic structure and transmission risk of Mycobacterium tuberculosis in Yichang city of China

Objective

The burden of both general and drug-resistant tuberculosis in rural areas is higher than that in urban areas in China. To characterize the genetic structure and transmission risk of Mycobacterium tuberculosis in rural China, we used whole genome sequencing to analyze clinical strains collected from patients in two counties of Yichang for three consecutive years.

Methods

From 2018 to 2020, sputum samples were collected for cultures from patients with suspected tuberculosis in Yidu and Zigui county, and DNA was extracted from the positive strains for genome sequencing. The online SAM-TB platform was used to identify the genotypes and drug resistance-related mutations of each strain, establish a phylogenetic tree, and calculated the genetic distances between pairwise strains. Twelve single nucleotide polymorphisms (SNPs) were used as thresholds to identify transmission clusters. The risk of related factors was estimated by univariable and multivariable logistic regression.

Results

A total of 161 out of the collected 231 positive strains were enrolled for analysis, excluding non-tuberculous mycobacterium and duplicate strains from the same patient. These strains belonged to Lineage 2 (92, 57.1%) and Lineage 4 (69, 42.9%), respectively. A total of 49 (30.4%) strains were detected with known drug resistance-related mutations, including 6 (3.7%) multidrug-resistant-TB (MDR-TB) strains and 11 (6.8%) RIF-resistant INH-susceptible TB (Rr-TB) strains. Six of the MDR/Rr-TB (35.3%) were also resistant to fluoroquinolones, which made them pre-extensively drug-resistant TB (pre-XDR-TB). There were another seven strains with mono-resistance to fluoroquinolones and one strain with resistance to both INH and fluoroquinolones, making the overall rate of fluoroquinolones resistance 8.7% (14/161). A total of 50 strains (31.1%) were identified as transmission clusters. Patients under 45 years old (adjusted odds ratio 3.46 [95% confidential intervals 1.28-9.35]), treatment-naive patients (6.14 [1.39-27.07]) and patients infected by lineage 4 strains (2.22 [1.00-4.91]) had a higher risk of transmission.

Conclusion

The drug resistance of tuberculosis in rural China, especially to the second-line drug fluoroquinolones, is relatively serious. The standardized treatment for patients and the clinical use of fluoroquinolones warrant attention. At the same time, the recent transmission risk of tuberculosis is high, and rapid diagnosis and treatment management at the primary care needs to be strengthened.

Genomic analysis of Mycobacterium brumae sustains its nonpathogenic and immunogenic phenotype

Mycobacterium brumae is a rapid-growing, non-pathogenic Mycobacterium species, originally isolated from environmental and human samples in Barcelona, Spain. Mycobacterium brumae is not pathogenic and it's in vitro phenotype and immunogenic properties have been well characterized. However, the knowledge of its underlying genetic composition is still incomplete. In this study, we first describe the 4 Mb genome of the M. brumae type strain ATCC 51384T assembling PacBio reads, and second, we assess the low intraspecies variability by comparing the type strain with Illumina reads from three additional strains. Mycobacterium brumae genome is composed of a circular chromosome with a high GC content of 69.2% and containing 3,791 CDSs, 97 pseudogenes, one prophage and no CRISPR loci. Mycobacterium brumae has shown no pathogenic potential in in vivo experiments, and our genomic analysis confirms its phylogenetic position with other non-pathogenic and rapid growing mycobacteria. Accordingly, we determined the absence of virulence-related genes, such as ESX-1 locus and most PE/PPE genes, among others. Although the immunogenic potential of M. brumae was proved to be as high as Mycobacterium bovis BCG, the only mycobacteria licensed to treat cancer, the genomic content of M. tuberculosis T cell and B cell antigens in M. brumae genome is considerably lower than those antigens present in M. bovis BCG genome. Overall, this work provides relevant genomic data on one of the species of the mycobacterial genus with high therapeutic potential.

Comparative genomics of drug-resistant strains of Mycobacterium tuberculosis in Ecuador

BACKGROUND

Tuberculosis is a serious infectious disease affecting millions of people. In spite of efforts to reduce the disease, increasing antibiotic resistance has contributed to persist in the top 10 causes of death worldwide. In fact, the increased cases of multi (MDR) and extreme drug resistance (XDR) worldwide remains the main challenge for tuberculosis control. Whole genome sequencing is a powerful tool for predicting drug resistance-related variants, studying lineages, tracking transmission, and defining outbreaks. This study presents the identification and characterization of resistant clinical isolates of Mycobacterium tuberculosis including a phylogenetic and molecular resistance profile study by sequencing the complete genome of 24 strains from different provinces of Ecuador.

RESULTS

Genomic sequencing was used to identify the variants causing resistance. A total of 15/21 isolates were identified as MDR, 4/21 as pre-XDR and 2/21 as XDR, with three isolates discarded due to low quality; the main sub-lineage was LAM (61.9%) and Haarlem (19%) but clades X, T and S were identified. Of the six pre-XDR and XDR strains, it is noteworthy that five come from females; four come from the LAM sub-lineage and two correspond to the X-class sub-lineage. A core genome of 3,750 genes, distributed in 295 subsystems, was determined. Among these, 64 proteins related to virulence and implicated in the pathogenicity of M. tuberculosis and 66 possible pharmacological targets stand out. Most variants result in nonsynonymous amino acid changes and the most frequent genotypes were identified as conferring resistance to rifampicin, isoniazid, ethambutol, para-aminosalicylic acid and streptomycin. However, an increase in the resistance to fluoroquinolones was detected.

CONCLUSION

This work shows for the first time the variability of circulating resistant strains between men and women in Ecuador, highlighting the usefulness of genomic sequencing for the identification of emerging resistance. In this regard, we found an increase in fluoroquinolone resistance. Further sampling effort is needed to determine the total variability and associations with the metadata obtained to generate better health policies.

TransFlow: a Snakemake workflow for transmission analysis of Mycobacterium tuberculosis whole-genome sequencing data

MOTIVATION

Whole-genome sequencing (WGS) is increasingly used to aid the understanding of Mycobacterium tuberculosis (MTB) transmission. The epidemiological analysis of tuberculosis based on the WGS technique requires a diverse collection of bioinformatics tools. Effectively using these analysis tools in a scalable and reproducible way can be challenging, especially for non-experts.

RESULTS

Here, we present TransFlow (Transmission Workflow), a user-friendly, fast, efficient and comprehensive WGS-based transmission analysis pipeline. TransFlow combines some state-of-the-art tools to take transmission analysis from raw sequencing data, through quality control, sequence alignment and variant calling, into downstream transmission clustering, transmission network reconstruction and transmission risk factor inference, together with summary statistics and data visualization in a summary report. TransFlow relies on Snakemake and Conda to resolve dependencies among consecutive processing steps and can be easily adapted to any computation environment.

AVAILABILITY AND IMPLEMENTATION

TransFlow is free available at https://github.com/cvn001/transflow.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Whole-Genome Sequencing for Resistance Level Prediction in Multidrug-Resistant Tuberculosis

Defining the precise relationship between resistance mutations and quantitative phenotypic drug susceptibility testing will increase the value of whole-genome sequencing (WGS) for predicting tuberculosis drug resistance. However, a large number of WGS data sets currently lack corresponding quantitative phenotypic data—the MICs. Using MYCOTBI plates, we determined the MICs to nine antituberculosis drugs for 154 clinical multidrug-resistant tuberculosis isolates from the Shenzhen Center for Chronic Disease Control in Shenzhen, China. Comparing MICs with predicted drug-resistance profiles inferred by WGS showed that WGS could predict the levels of resistance to isoniazid, rifampicin, streptomycin, fluoroquinolones, and aminoglycosides. We also found some mutations that may not be associated with drug resistance, such as EmbB D328G, mutations in the gid gene, and C−12T in the eis promoter. However, some strains carrying the same mutations showed different levels of resistance to the corresponding drugs. The MICs of different strains with the RpsL K88R, fabG1 C−15T mutations and some with mutations in embB and rpoB, had MICs to the corresponding drugs that varied by 8-fold or more. This variation is unexplained but could be influenced by the bacterial genetic background. Additionally, we found that 32.3% of rifampicin-resistant isolates were rifabutin-susceptible, particularly those with rpoB mutations H445D, H445L, H445S, D435V, D435F, L452P, S441Q, and S441V. Studying the influence of bacterial genetic background on the MIC and the relationship between rifampicin-resistant mutations and rifabutin resistance levels should improve the ability of WGS to guide the selection of medical treatment regimens.

IMPORTANCE Whole-genome sequencing (WGS) has excellent potential in drug-resistance prediction. The MICs are essential indications of adding a particular antituberculosis drug dosage or changing the entire treatment regimen. However, the relationship between many known drug-resistant mutations and MICs is unclear, especially for rarer ones. The results showed that WGS could predict resistance levels to isoniazid, rifampicin, streptomycin, fluoroquinolones, and aminoglycosides. However, some mutations may not be associated with drug resistance, and some others may confer various MICs to strains carrying them. Also, 32.3% of rifampicin (RIF)-resistant strains were classified as sensitive to rifabutin (RFB), and some mutations in the rpoB gene may be associated with this phenotype. Our data on the MIC distribution of strains with some rarer mutations add to the accumulated data on the resistance level associated with such mutations to help guide further research and draw meaningful conclusions.