GenTB: A user-friendly genome-based predictor for tuberculosis resistance powered by machine learning

Drug-resistant tuberculosis: a persistent global health concern

Drug-resistant tuberculosis (TB) is estimated to cause 13% of all antimicrobial resistance-attributable deaths worldwide and is driven by both ongoing resistance acquisition and person-to-person transmission. Poor outcomes are exacerbated by late diagnosis and inadequate access to effective treatment. Advances in rapid molecular testing have recently improved the diagnosis of TB and drug resistance. Next-generation sequencing of Mycobacterium tuberculosis has increased our understanding of genetic resistance mechanisms and can now detect mutations associated with resistance phenotypes. All-oral, shorter drug regimens that can achieve high cure rates of drug-resistant TB within 6-9 months are now available and recommended but have yet to be scaled to global clinical use. Promising regimens for the prevention of drug-resistant TB among high-risk contacts are supported by early clinical trial data but final results are pending. A person-centred approach is crucial in managing drug-resistant TB to reduce the risk of poor treatment outcomes, side effects, stigma and mental health burden associated with the diagnosis. In this Review, we describe current surveillance of drug-resistant TB and the causes, risk factors and determinants of drug resistance as well as the stigma and mental health considerations associated with it. We discuss recent advances in diagnostics and drug-susceptibility testing and outline the progress in developing better treatment and preventive therapies.

Detection of multidrug-resistance in Mycobacterium tuberculosis by phenotype- and molecular-based assays

BACKGROUND

The whole-genome sequencing (WGS) is becoming an increasingly effective tool for rapid and accurate detection of drug resistance in Mycobacterium tuberculosis complex (MTBC). This approach, however, has still been poorly evaluated on strains from Central and Eastern European countries. The purpose of this study was to assess the performance of WGS against conventional drug susceptibility testing (DST) for the detection of multi-drug resistant (MDR) phenotypes among MTBC clinical strains from Poland and Lithuania.

METHODS

The study included 208 MTBC strains (130 MDR; 78 drug susceptible), recovered from as many tuberculosis patients in Lithuania and Poland between 2018 and 2021. Resistance to rifampicin (RIF) and isoniazid (INH) was assessed by Critical Concentration (CC) and Minimum Inhibitory Concentration (MIC) DST as well as molecular-based techniques, including line-probe assay (LPA) and WGS. The analysis of WGS results was performed using bioinformatic pipeline- and software-based tools.

RESULTS

The results obtained with the CC DST were more congruent with those by LPA compared to pipeline-based WGS. Software-based tools showed excellent concordance with pipeline-based analysis in prediction of RIF/INH resistance. The RIF-resistant strains demonstrated a relatively homogenous MIC distribution with the mode at the highest tested MIC value. The most frequent RIF-resistance conferring mutation was rpoB S450L. The mode MIC for INH was two-fold higher among double katG and inhA mutants than among single katG mutants. The overall rate of discordant results between all methods was calculated at 5.3%. Three strains had discordant results by both genotypic methods (LPA and pipeline-based WGS), one strain by LPA only, three strains by MIC DST, two strains by both MIC DST and pipeline-based WGS, and the remaining two strains showed discordant results with all three methods, compared to CC DST.

CONCLUSIONS

Considering MIC DST results, current CCs of the first-line anti-TB drugs might be inappropriately high and may need to be revised. Both molecular methods demonstrated 100% specificity, while pipeline-based WGS had slightly lower sensitivity for RIF and INH than LPA, compared to CC DST.

AI-Driven Thoracic X-ray Diagnostics: Transformative Transfer Learning for Clinical Validation in Pulmonary Radiography

Our research evaluates advanced artificial (AI) methodologies to enhance diagnostic accuracy in pulmonary radiography. Utilizing DenseNet121 and ResNet50, we analyzed 108,948 chest X-ray images from 32,717 patients and DenseNet121 achieved an area under the curve (AUC) of 94% in identifying the conditions of pneumothorax and oedema. The model's performance surpassed that of expert radiologists, though further improvements are necessary for diagnosing complex conditions such as emphysema, effusion, and hernia. Clinical validation integrating Latent Dirichlet Allocation (LDA) and Named Entity Recognition (NER) demonstrated the potential of natural language processing (NLP) in clinical workflows. The NER system achieved a precision of 92% and a recall of 88%. Sentiment analysis using DistilBERT provided a nuanced understanding of clinical notes, which is essential for refining diagnostic decisions. XGBoost and SHapley Additive exPlanations (SHAP) enhanced feature extraction and model interpretability. Local Interpretable Model-agnostic Explanations (LIME) and occlusion sensitivity analysis further enriched transparency, enabling healthcare providers to trust AI predictions. These AI techniques reduced processing times by 60% and annotation errors by 75%, setting a new benchmark for efficiency in thoracic diagnostics. The research explored the transformative potential of AI in medical imaging, advancing traditional diagnostics and accelerating medical evaluations in clinical settings.

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.

Pan-lineage Mycobacterium tuberculosis reference genome for enhanced molecular diagnosis

In Mycobacterium tuberculosis (MTB) control, whole genome sequencing-based molecular drug susceptibility testing (molDST-WGS) has emerged as a pivotal tool. However, the current reliance on a single-strain reference limits molDST-WGS's true potential. To address this, we introduce a new pan-lineage reference genome, 'MtbRf'. We assembled 'unmapped' reads from 3,614 MTB genomes (751 L1; 881 L2; 1,700 L3; and 282 L4) into 35 shared, annotated contigs (54 coding sequences [CDSs]). We constructed MtbRf through: (1) searching for contig homologues among genome database that precipitate results uniquely within Mycobacteria genus; (2) comparing genomes with H37Rv ('lift-over') to define 18 insertions; and (3) filling gaps in H37Rv with insertions. MtbRf adds 1.18% sequences to H37rv, salvaging >60% of previously unmapped reads. Transcriptomics confirmed gene expression of new CDSs. The new variants provided a moderate DST predictive value (AUROC 0.60-0.75). MtbRf thus unveils previously hidden genomic information and lays the foundation for lineage-specific molDST-WGS.

Surveillance of fluoroquinolones resistance in rifampicin-susceptible tuberculosis in eastern China with whole-genome sequencing-based approach

Background

Leveraging well-established DNA-level drug resistance mechanisms, whole-genome sequencing (WGS) has emerged as a valuable methodology for predicting drug resistance. As the most effective second-line anti-tuberculosis (anti-TB) drugs, fluoroquinoloness (FQs) are generally used to treat multidrug-resistant tuberculosis (MDR-TB, defined as being resistant to resistant to rifampicin and isoniazid) or rifampicin-resistant tuberculosis (RR-TB). However, FQs are also commonly used in the management of other bacterial infections. There are few published data on the rates of FQs resistance among rifampicin-susceptible TB. The prevalence of FQs resistance among TB patients who are rifampicin-susceptible has not been studied in Zhejiang Province, China. The goal of this study was to provide a baseline characterization of the prevalence of FQs resistance, particularly among rifampicin-susceptible TB in Zhejiang Province, China.

Methods

Based on WGS, we have investigated the prevalence of FQs resistance among rifampicin-susceptible TB in Zhejiang Province. All pulmonary TB patients with positive cultures who were identified in Zhejiang area during TB drug resistance surveillance from 2018 to 2019 have enrolled in this population-based retrospective study.

Results

The rate of FQs resistance was 4.6% (32/698) among TB, 4.0% (27/676) among rifampicin-susceptible TB, and 22.7% (5/22) among RR-TB. According to WGS, strains that differ within 12 single-nucleotide polymorphisms (SNPs) were considered to be transmission of FQ-resistant strains. Specifically, 3.7% (1/27) of FQs resistance was caused by the transmission of FQs-resistant strains among the rifampicin-susceptible TB and 40.7% (11/27) of FQs resistance was identified as hetero-resistance.

Conclusion

The prevalence of FQs resistance among TB patients who were rifampicin-susceptible was severe in Zhejiang. The emergence of FQs resistance in TB isolates that are rifampicin-susceptible was mainly caused by the selection of drug-resistant strains. In order to prevent the emergence of FQs resistance, the WGS-based surveillance system for TB should be urgently established, and clinical awareness of the responsible use of FQs for respiratory infections should be enhanced.

Scalable de novo classification of antibiotic resistance of Mycobacterium tuberculosis

MOTIVATION

World Health Organization estimates that there were over 10 million cases of tuberculosis (TB) worldwide in 2019, resulting in over 1.4 million deaths, with a worrisome increasing trend yearly. The disease is caused by Mycobacterium tuberculosis (MTB) through airborne transmission. Treatment of TB is estimated to be 85% successful, however, this drops to 57% if MTB exhibits multiple antimicrobial resistance (AMR), for which fewer treatment options are available.

RESULTS

We develop a robust machine-learning classifier using both linear and nonlinear models (i.e. LASSO logistic regression (LR) and random forests (RF)) to predict the phenotypic resistance of Mycobacterium tuberculosis (MTB) for a broad range of antibiotic drugs. We use data from the CRyPTIC consortium to train our classifier, which consists of whole genome sequencing and antibiotic susceptibility testing (AST) phenotypic data for 13 different antibiotics. To train our model, we assemble the sequence data into genomic contigs, identify all unique 31-mers in the set of contigs, and build a feature matrix M, where M[i, j] is equal to the number of times the ith 31-mer occurs in the jth genome. Due to the size of this feature matrix (over 350 million unique 31-mers), we build and use a sparse matrix representation. Our method, which we refer to as MTB++, leverages compact data structures and iterative methods to allow for the screening of all the 31-mers in the development of both LASSO LR and RF. MTB++ is able to achieve high discrimination (F-1 >80%) for the first-line antibiotics. Moreover, MTB++ had the highest F-1 score in all but three classes and was the most comprehensive since it had an F-1 score >75% in all but four (rare) antibiotic drugs. We use our feature selection to contextualize the 31-mers that are used for the prediction of phenotypic resistance, leading to some insights about sequence similarity to genes in MEGARes. Lastly, we give an estimate of the amount of data that is needed in order to provide accurate predictions.

AVAILABILITY

The models and source code are publicly available on Github at https://github.com/M-Serajian/MTB-Pipeline.

Combating antimicrobial resistance in malaria, HIV and tuberculosis

Antimicrobial resistance poses a significant threat to the sustainability of effective treatments against the three most prevalent infectious diseases: malaria, human immunodeficiency virus (HIV) infection and tuberculosis. Therefore, there is an urgent need to develop novel drugs and treatment protocols capable of reducing the emergence of resistance and combating it when it does occur. In this Review, we present an overview of the status and underlying molecular mechanisms of drug resistance in these three diseases. We also discuss current strategies to address resistance during the research and development of next-generation therapies. These strategies vary depending on the infectious agent and the array of resistance mechanisms involved. Furthermore, we explore the potential for cross-fertilization of knowledge and technology among these diseases to create innovative approaches for minimizing drug resistance and advancing the discovery and development of new anti-infective treatments. In conclusion, we advocate for the implementation of well-defined strategies to effectively mitigate and manage resistance in all interventions against infectious diseases.

Drug resistance and epidemiological success of modern Mycobacterium tuberculosis lineages in western India

BACKGROUND

Transmission is contributing to the slow decline of tuberculosis (TB) incidence globally. Drivers of TB transmission in India, the country estimated to carry a quarter of the World's burden, are not well studied. We conducted a genomic epidemiology study to compare epidemiological success, host factors and drug resistance (DR) among the four major Mycobacterium tuberculosis (Mtb) lineages (L1-4) circulating in Pune, India.

METHODS

We performed whole-genome sequencing (WGS) of Mtb sputum culture-positive isolates from participants in two prospective cohort studies and predicted genotypic susceptibility using a validated random forest model. We used maximum likelihood estimation to build phylogenies. We compared lineage specific phylogenetic and time-scaled metrics to assess epidemiological success.

RESULTS

Of the 642 isolates that underwent WGS, 612 met sequence quality criteria. Most isolates belonged to L3 (44.6%). The majority (61.1%) of multidrug-resistant isolates belonged to L2 (P < 0.001). In molecular dating, L2 demonstrated a higher rate and more recent resistance acquisition. We measured higher clustering, and time-scaled haplotypic density (THD) for L4 and L2 compared to L3 and/or L1 suggesting higher epidemiological success. L4 demonstrated higher THD and clustering (OR 5.1 (95% CI 2.3-12.3) in multivariate models controlling for host factors and DR.

CONCLUSION

L2 shows a higher frequency of DR and both L2 and L4 demonstrate evidence of higher epidemiological success than L3 or L1 in the study setting. Our findings highlight the need for contact tracing around TB cases, and heightened surveillance of TB DR in India.

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.

Tackling the Antimicrobial Resistance "Pandemic" with Machine Learning Tools: A Summary of Available Evidence

Antimicrobial resistance is recognised as one of the top threats healthcare is bound to face in the future. There have been various attempts to preserve the efficacy of existing antimicrobials, develop new and efficient antimicrobials, manage infections with multi-drug resistant strains, and improve patient outcomes, resulting in a growing mass of routinely available data, including electronic health records and microbiological information that can be employed to develop individualised antimicrobial stewardship. Machine learning methods have been developed to predict antimicrobial resistance from whole-genome sequencing data, forecast medication susceptibility, recognise epidemic patterns for surveillance purposes, or propose new antibacterial treatments and accelerate scientific discovery. Unfortunately, there is an evident gap between the number of machine learning applications in science and the effective implementation of these systems. This narrative review highlights some of the outstanding opportunities that machine learning offers when applied in research related to antimicrobial resistance. In the future, machine learning tools may prove to be superbugs' kryptonite. This review aims to provide an overview of available publications to aid researchers that are looking to expand their work with new approaches and to acquaint them with the current application of machine learning techniques in this field.

Estimation of country-specific tuberculosis resistance antibiograms using pathogen genomics and machine learning

BACKGROUND

Global tuberculosis (TB) drug resistance (DR) surveillance focuses on rifampicin. We examined the potential of public and surveillance Mycobacterium tuberculosis (Mtb) whole-genome sequencing (WGS) data, to generate expanded country-level resistance prevalence estimates (antibiograms) using in silico resistance prediction.

METHODS

We curated and quality-controlled Mtb WGS data. We used a validated random forest model to predict phenotypic resistance to 12 drugs and bias-corrected for model performance, outbreak sampling and rifampicin resistance oversampling. Validation leveraged a national DR survey conducted in South Africa.

RESULTS

Mtb isolates from 29 countries (n=19 149) met sequence quality criteria. Global marginal genotypic resistance among mono-resistant TB estimates overlapped with the South African DR survey, except for isoniazid, ethionamide and second-line injectables, which were underestimated (n=3134). Among multidrug resistant (MDR) TB (n=268), estimates overlapped for the fluoroquinolones but overestimated other drugs. Globally pooled mono-resistance to isoniazid was 10.9% (95% CI: 10.2-11.7%, n=14 012). Mono-levofloxacin resistance rates were highest in South Asia (Pakistan 3.4% (0.1-11%), n=111 and India 2.8% (0.08-9.4%), n=114). Given the recent interest in drugs enhancing ethionamide activity and their expected activity against isolates with resistance discordance between isoniazid and ethionamide, we measured this rate and found it to be high at 74.4% (IQR: 64.5-79.7%) of isoniazid-resistant isolates predicted to be ethionamide susceptible. The global susceptibility rate to pyrazinamide and levofloxacin among MDR was 15.1% (95% CI: 10.2-19.9%, n=3964).

CONCLUSIONS

This is the first attempt at global Mtb antibiogram estimation. DR prevalence in Mtb can be reliably estimated using public WGS and phenotypic resistance prediction for key antibiotics, but public WGS data demonstrates oversampling of isolates with higher resistance levels than MDR. Nevertheless, our results raise concerns about the empiric use of short-course fluoroquinolone regimens for drug-susceptible TB in South Asia and indicate underutilisation of ethionamide in MDR treatment.

Assessing computational predictions of antimicrobial resistance phenotypes from microbial genomes

The advent of rapid whole-genome sequencing has created new opportunities for computational prediction of antimicrobial resistance (AMR) phenotypes from genomic data. Both rule-based and machine learning (ML) approaches have been explored for this task, but systematic benchmarking is still needed. Here, we evaluated four state-of-the-art ML methods (Kover, PhenotypeSeeker, Seq2Geno2Pheno and Aytan-Aktug), an ML baseline and the rule-based ResFinder by training and testing each of them across 78 species-antibiotic datasets, using a rigorous benchmarking workflow that integrates three evaluation approaches, each paired with three distinct sample splitting methods. Our analysis revealed considerable variation in the performance across techniques and datasets. Whereas ML methods generally excelled for closely related strains, ResFinder excelled for handling divergent genomes. Overall, Kover most frequently ranked top among the ML approaches, followed by PhenotypeSeeker and Seq2Geno2Pheno. AMR phenotypes for antibiotic classes such as macrolides and sulfonamides were predicted with the highest accuracies. The quality of predictions varied substantially across species-antibiotic combinations, particularly for beta-lactams; across species, resistance phenotyping of the beta-lactams compound, aztreonam, amoxicillin/clavulanic acid, cefoxitin, ceftazidime and piperacillin/tazobactam, alongside tetracyclines demonstrated more variable performance than the other benchmarked antibiotics. By organism, Campylobacter jejuni and Enterococcus faecium phenotypes were more robustly predicted than those of Escherichia coli, Staphylococcus aureus, Salmonella enterica, Neisseria gonorrhoeae, Klebsiella pneumoniae, Pseudomonas aeruginosa, Acinetobacter baumannii, Streptococcus pneumoniae and Mycobacterium tuberculosis. In addition, our study provides software recommendations for each species-antibiotic combination. It furthermore highlights the need for optimization for robust clinical applications, particularly for strains that diverge substantially from those used for training.

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.

Clinical Whole-Genome Sequencing Assay for Rapid Mycobacterium tuberculosis Complex First-Line Drug Susceptibility Testing and Phylogenetic Relatedness Analysis

The global rise of drug resistant tuberculosis has highlighted the need for improved diagnostic technologies that provide rapid and reliable drug resistance results. Here, we develop and validate a whole genome sequencing (WGS)-based test for identification of mycobacterium tuberculosis complex (MTB) drug resistance to rifampin, isoniazid, pyrazinamide, ethambutol, and streptomycin. Through comparative analysis of drug resistance results from WGS-based testing and phenotypic drug susceptibility testing (DST) of 38 clinical MTB isolates from patients receiving care in Los Angeles, CA, we found an overall concordance between methods of 97.4% with equivalent performance across culture media. Critically, prospective analysis of 11 isolates showed that WGS-based testing provides results an average of 36 days faster than phenotypic culture-based methods. We showcase the additional benefits of WGS data by investigating a suspected laboratory contamination event and using phylogenetic analysis to search for cryptic local transmission, finding no evidence of community spread amongst our patient population in the past six years. WGS-based testing for MTB drug resistance has the potential to greatly improve diagnosis of drug resistant MTB by accelerating turnaround time while maintaining accuracy and providing additional benefits for infection control, lab safety, and public health applications.

Drug resistance prediction for Mycobacterium tuberculosis with reference graphs

Tuberculosis is a global pandemic disease with a rising burden of antimicrobial resistance. As a result, the World Health Organization (WHO) has a goal of enabling universal access to drug susceptibility testing (DST). Given the slowness of and infrastructure requirements for phenotypic DST, whole-genome sequencing, followed by genotype-based prediction of DST, now provides a route to achieving this. Since a central component of genotypic DST is to detect the presence of any known resistance-causing mutations, a natural approach is to use a reference graph that allows encoding of known variation. We have developed DrPRG (Drug resistance Prediction with Reference Graphs) using the bacterial reference graph method Pandora. First, we outline the construction of a Mycobacterium tuberculosis drug resistance reference graph. The graph is built from a global dataset of isolates with varying drug susceptibility profiles, thus capturing common and rare resistance- and susceptible-associated haplotypes. We benchmark DrPRG against the existing graph-based tool Mykrobe and the haplotype-based approach of TBProfiler using 44 709 and 138 publicly available Illumina and Nanopore samples with associated phenotypes. We find that DrPRG has significantly improved sensitivity and specificity for some drugs compared to these tools, with no significant decreases. It uses significantly less computational memory than both tools, and provides significantly faster runtimes, except when runtime is compared to Mykrobe with Nanopore data. We discover and discuss novel insights into resistance-conferring variation for M. tuberculosis - including deletion of genes katG and pncA - and suggest mutations that may warrant reclassification as associated with resistance.

Machine Learning of the Whole Genome Sequence of Mycobacterium tuberculosis: A Scoping PRISMA-Based Review

Tuberculosis (TB) remains one of the most significant global health problems, posing a significant challenge to public health systems worldwide. However, diagnosing drug-resistant tuberculosis (DR-TB) has become increasingly challenging due to the rising number of multidrug-resistant (MDR-TB) cases, despite the development of new TB diagnostic tools. Even the World Health Organization-recommended methods such as Xpert MTB/XDR or Truenat are unable to detect all the Mycobacterium tuberculosis genome mutations associated with drug resistance. While Whole Genome Sequencing offers a more precise DR profile, the lack of user-friendly bioinformatics analysis applications hinders its widespread use. This review focuses on exploring various artificial intelligence models for predicting DR-TB profiles, analyzing relevant English-language articles using the PRISMA methodology through the Covidence platform. Our findings indicate that an Artificial Neural Network is the most commonly employed method, with non-statistical dimensionality reduction techniques preferred over traditional statistical approaches such as Principal Component Analysis or t-distributed Stochastic Neighbor Embedding.

Phase variation as a major mechanism of adaptation in Mycobacterium tuberculosis complex

Phase variation induced by insertions and deletions (INDELs) in genomic homopolymeric tracts (HT) can silence and regulate genes in pathogenic bacteria, but this process is not characterized in MTBC (Mycobacterium tuberculosis complex) adaptation. We leverage 31,428 diverse clinical isolates to identify genomic regions including phase-variants under positive selection. Of 87,651 INDEL events that emerge repeatedly across the phylogeny, 12.4% are phase-variants within HTs (0.02% of the genome by length). We estimated the in-vitro frameshift rate in a neutral HT at 100× the neutral substitution rate at [Formula: see text] frameshifts/HT/year. Using neutral evolution simulations, we identified 4,098 substitutions and 45 phase-variants to be putatively adaptive to MTBC (P < 0.002). We experimentally confirm that a putatively adaptive phase-variant alters the expression of espA, a critical mediator of ESX-1-dependent virulence. Our evidence supports the hypothesis that phase variation in the ESX-1 system of MTBC can act as a toggle between antigenicity and survival in the host.

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.

Machine Learning for Antimicrobial Resistance Prediction: Current Practice, Limitations, and Clinical Perspective

Antimicrobial resistance (AMR) is a global health crisis that poses a great threat to modern medicine. Effective prevention strategies are urgently required to slow the emergence and further dissemination of AMR. Given the availability of data sets encompassing hundreds or thousands of pathogen genomes, machine learning (ML) is increasingly being used to predict resistance to different antibiotics in pathogens based on gene content and genome composition. A key objective of this work is to advocate for the incorporation of ML into front-line settings but also highlight the further refinements that are necessary to safely and confidently incorporate these methods. The question of what to predict is not trivial given the existence of different quantitative and qualitative laboratory measures of AMR. ML models typically treat genes as independent predictors, with no consideration of structural and functional linkages; they also may not be accurate when new mutational variants of known AMR genes emerge. Finally, to have the technology trusted by end users in public health settings, ML models need to be transparent and explainable to ensure that the basis for prediction is clear. We strongly advocate that the next set of AMR-ML studies should focus on the refinement of these limitations to be able to bridge the gap to diagnostic implementation.

Whole-genome sequencing for surveillance of fluoroquinolone resistance in rifampicin-susceptible tuberculosis in a rural district of Shanghai: A 10-year retrospective study

Background

Fluoroquinolones (FQs) are the most important second-line anti-tuberculosis (anti-TB) drugs, primarily used for the treatment of multidrug- or rifampicin-resistant TB (MDR/RR-TB). However, FQs are also commonly used to treat other bacterial infections. There are few published data on the rates of FQ resistance among rifampicin-susceptible TB.

Methods

We used whole-genome sequencing (WGS) to determine the prevalence of FQ resistance among rifampicin-susceptible TB in a rural district of Shanghai. This was a population-based retrospective study of all culture-positive pulmonary TB patients diagnosed in the Chongming district of Shanghai, China during 2009-2018.

Results

The rate of FQ resistance was 8.4% (29/345) among TB, 6.2% (20/324) among rifampicin-susceptible TB, and 42.9% (9/21) among MDR/RR-TB. Transmission of FQ-resistant strains was defined as strains differing within 12 single-nucleotide polymorphisms (SNPs) based on WGS. Among the rifampicin-susceptible TB, 20% (4/20) of FQ resistance was caused by the transmission of FQ-resistant strains and 45% (9/20) of FQ resistance was identified as hetero-resistance.

Conclusions

The prevalence of FQ resistance in rifampicin-susceptible TB was higher than expected in Shanghai. Both the transmission and the selection of drug-resistant strains drive the emergence of FQ resistance in rifampicin-susceptible TB isolates. Therefore, the WGS-based surveillance system for TB should be urgently established and the clinical awareness of the rational use of FQs for respiratory infections should be enhanced to prevent the premature occurrence of FQ resistance.

Knowledge structure and emerging trends in the application of deep learning in genetics research: A bibliometric analysis [2000–2021]

Introduction: Deep learning technology has been widely used in genetic research because of its characteristics of computability, statistical analysis, and predictability. Herein, we aimed to summarize standardized knowledge and potentially innovative approaches for deep learning applications of genetics by evaluating publications to encourage more research.

Methods: The Science Citation Index Expanded TM (SCIE) database was searched for deep learning applications for genomics-related publications. Original articles and reviews were considered. In this study, we derived a clustered network from 69,806 references that were cited by the 1,754 related manuscripts identified. We used CiteSpace and VOSviewer to identify countries, institutions, journals, co-cited references, keywords, subject evolution, path, current characteristics, and emerging topics.

Results: We assessed the rapidly increasing publications concerned about deep learning applications of genomics approaches and identified 1,754 articles that published reports focusing on this subject. Among these, a total of 101 countries and 2,487 institutes contributed publications, The United States of America had the most publications (728/1754) and the highest h-index, and the US has been in close collaborations with China and Germany. The reference clusters of SCI articles were clustered into seven categories: deep learning, logic regression, variant prioritization, random forests, scRNA-seq (single-cell RNA-seq), genomic regulation, and recombination. The keywords representing the research frontiers by year were prediction (2016–2021), sequence (2017–2021), mutation (2017–2021), and cancer (2019–2021).

Conclusion: Here, we summarized the current literature related to the status of deep learning for genetics applications and analyzed the current research characteristics and future trajectories in this field. This work aims to provide resources for possible further intensive exploration and encourages more researchers to overcome the research of deep learning applications in genetics.

The Application of Artificial Intelligence in the Diagnosis and Drug Resistance Prediction of Pulmonary Tuberculosis

With the increasing incidence and mortality of pulmonary tuberculosis, in addition to tough and controversial disease management, time-wasting and resource-limited conventional approaches to the diagnosis and differential diagnosis of tuberculosis are still awkward issues, especially in countries with high tuberculosis burden and backwardness. In the meantime, the climbing proportion of drug-resistant tuberculosis poses a significant hazard to public health. Thus, auxiliary diagnostic tools with higher efficiency and accuracy are urgently required. Artificial intelligence (AI), which is not new but has recently grown in popularity, provides researchers with opportunities and technical underpinnings to develop novel, precise, rapid, and automated implements for pulmonary tuberculosis care, including but not limited to tuberculosis detection. In this review, we aimed to introduce representative AI methods, focusing on deep learning and radiomics, followed by definite descriptions of the state-of-the-art AI models developed using medical images and genetic data to detect pulmonary tuberculosis, distinguish the infection from other pulmonary diseases, and identify drug resistance of tuberculosis, with the purpose of assisting physicians in deciding the appropriate therapeutic schedule in the early stage of the disease. We also enumerated the challenges in maximizing the impact of AI in this field such as generalization and clinical utility of the deep learning models.

A convolutional neural network highlights mutations relevant to antimicrobial resistance in Mycobacterium tuberculosis

Long diagnostic wait times hinder international efforts to address antibiotic resistance in M. tuberculosis. Pathogen whole genome sequencing, coupled with statistical and machine learning models, offers a promising solution. However, generalizability and clinical adoption have been limited by a lack of interpretability, especially in deep learning methods. Here, we present two deep convolutional neural networks that predict antibiotic resistance phenotypes of M. tuberculosis isolates: a multi-drug CNN (MD-CNN), that predicts resistance to 13 antibiotics based on 18 genomic loci, with AUCs 82.6-99.5% and higher sensitivity than state-of-the-art methods; and a set of 13 single-drug CNNs (SD-CNN) with AUCs 80.1-97.1% and higher specificity than the previous state-of-the-art. Using saliency methods to evaluate the contribution of input sequence features to the SD-CNN predictions, we identify 18 sites in the genome not previously associated with resistance. The CNN models permit functional variant discovery, biologically meaningful interpretation, and clinical applicability.

Whole-Genome Sequencing Has the Potential To Improve Treatment for Rifampicin-Resistant Tuberculosis in High-Burden Settings: a Retrospective Cohort Study

Treatment of multidrug-resistant or rifampicin-resistant tuberculosis (MDR/RR-TB), although improved in recent years with shorter, more tolerable regimens, remains largely standardized and based on limited drug susceptibility testing (DST). More individualized treatment with expanded DST access is likely to improve patient outcomes. To assess the potential of TB drug resistance prediction based on whole-genome sequencing (WGS) to provide more effective treatment regimens, we applied current South African treatment recommendations to a retrospective cohort of MDR/RR-TB patients from Khayelitsha, Cape Town. Routine DST and clinical data were used to retrospectively categorize patients into a recommended regimen, either a standardized short regimen or a longer individualized regimen. Potential regimen changes were then described with the addition of WGS-derived DST. WGS data were available for 1274 MDR/RR-TB patient treatment episodes across 2008 to 2017. Among 834 patients initially eligible for the shorter regimen, 385 (46%) may have benefited from reduced drug dosage or removing ineffective drugs when WGS data were considered. A further 187 (22%) patients may have benefited from more effective adjusted regimens. Among 440 patients initially eligible for a longer individualized regimen, 153 (35%) could have been switched to the short regimen. Overall, 305 (24%) patients had MDR/RR-TB with second-line TB drug resistance, where the availability of WGS-derived DST would have allowed more effective treatment individualization. These data suggest considerable benefits could accrue from routine access to WGS-derived resistance prediction. Advances in culture-free sequencing and expansion of the reference resistance mutation catalogue will increase the utility of WGS resistance prediction.

Advancing TB research using digitized programmatic data

The use of real-world data from national TB care programs has great potential to answer key research questions in TB control and is now opportune due to increasing digital data collection and storage. We summarize an expert stakeholder workshop conducted on this topic in October 2019, with perspectives from academics, national TB program officers, and data managers. We discuss challenges and opportunities in the use of TB programmatic data for research and describe digital data availability in two large, high TB burden countries, Brazil and South Africa. From this, we posit that with a standardized data collection set, improved data management, and greater collaboration, more TB programmatic data can be used for research with measurable public health impact.