Whole-genome sequencing mycobacteria susceptibility

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-08-21

Quick Answer

• Whole-genome sequencing (WGS) predicts Mycobacterium tuberculosis complex drug susceptibility by detecting resistance-associated mutations, with published sensitivities of 100% for rifampicin and isoniazid and 93.8% for pyrazinamide when compared against phenotypic testing.
• Laboratories considering WGS must validate their own DNA-library preparation workflows, bioinformatics pipelines, and quality-control parameters because manufacturer protocols require customization for mycobacterial genomes.
• WGS can correct phenotypic drug susceptibility testing errors, but discordant results require confirmatory testing and clinical correlation before treatment decisions are changed.

At a Glance

Understanding Whole-Genome Sequencing for Mycobacterial Susceptibility Testing

The Clinical Problem That Drives Interest in WGS

Tuberculosis remains a leading infectious cause of death worldwide, and antimicrobial resistance threatens treatment success across many regions. The World Health Organization identifies antimicrobial resistance as a global public health concern requiring coordinated surveillance and stewardship efforts. For Mycobacterium tuberculosis complex (MTBC) isolates, conventional phenotypic drug susceptibility testing (DST) has served as the reference method for decades, but it carries a fundamental limitation: the organism grows slowly, so results can take weeks to become available. During that interval, a patient with drug-resistant tuberculosis may receive ineffective therapy, allowing disease progression and further transmission.

Whole-genome sequencing offers a fundamentally different approach. Instead of observing whether a bacterial culture grows in the presence of a drug, WGS reads the complete genetic code of the isolate and searches for mutations known to confer resistance. This genotypic approach can produce results in days instead of weeks, and it provides information about multiple drugs simultaneously from a single sequencing run. The question facing laboratory professionals is whether WGS can replace phenotypic DST or whether it should serve as a complementary tool with specific strengths and limitations.

How WGS Predicts Drug Susceptibility

The logic underlying WGS-based susceptibility testing rests on a straightforward premise: specific genetic mutations cause specific resistance phenotypes. For Mycobacterium tuberculosis, decades of research have established catalogues of mutations associated with resistance to first-line and second-line drugs. When a laboratory sequences an isolate, bioinformatics software compares the detected mutations against these catalogues and generates a predicted susceptibility profile.

The most clinically important resistance mutations are well characterized. Mutations in the rpoB gene cause rifampicin resistance, with the S450L substitution being the most common variant identified in clinical isolates. Isoniazid resistance arises primarily through mutations in katG, particularly the S315T substitution, as well as through mutations in the inhA promoter region and other genes. Streptomycin resistance involves mutations in rpsL and rrs, while amikacin resistance is associated with specific rrs mutations. Ethambutol resistance commonly involves embB mutations, and pyrazinamide resistance is linked to pncA mutations.

A study from Papua, Indonesia, demonstrated the practical application of this approach. Researchers sequenced clinical isolates and identified mutations in rpoB including S450L, D435Y, H445Y, L430P, and Q432K that reduced rifampicin effectiveness. They also found katG S315T, kasA 312S, inhA I21V, and Rv1482c-fabG1 C-15T mutations contributing to isoniazid resistance, along with rpsL K43R and rrs A514C and A514T mutations for streptomycin resistance and rrs A514C for amikacin resistance. This study illustrates how WGS can simultaneously detect resistance determinants for multiple drugs from a single clinical isolate.

The DNA Library as the Critical First Step

Before sequencing can occur, the laboratory must prepare a DNA library. This process converts the genomic DNA extracted from a mycobacterial culture into a collection of short DNA fragments with attached adapters that allow the sequencing instrument to read them. The library represents the genomic content of the sample and must accurately reflect the original isolate for downstream predictions to be valid.

A detailed workflow published in Methods in Molecular Biology emphasizes that DNA-library preparation is the critical first step in the WGS process. Although commercial kits come with manufacturer instructions, the published protocol notes that details of the entire process, including quality controls, instrument parameters, and run evaluations, often need to be developed and customized by each laboratory. The authors provide an adapted Illumina protocol optimized for reduced reagent costs, acknowledging that successful WGS implementation requires laboratory-specific optimization instead of blind adherence to manufacturer defaults.

This point carries practical weight for laboratories considering WGS adoption. The quality of the DNA library directly determines the quality of the sequencing data, which in turn determines the reliability of resistance predictions. Laboratories must establish their own quality-control thresholds for DNA quantity, fragment size distribution, and library concentration. They must also validate their sequencing runs by monitoring metrics such as read depth, coverage uniformity, and error rates.

Evidence for WGS Performance in Routine Practice

A French Routine Laboratory Comparison

The strongest evidence for WGS performance in routine clinical practice comes from a study conducted at Lyon University Hospital in France. Researchers compared WGS-based prediction against phenotypic DST for all MTB isolates cultured during routine practice between November 2016 and July 2019, a total of 274 isolates. They tested first-line drugs using the Mycobacteria Growth Indicator Tube system for phenotypic DST and used both line probe assays (LPA) and WGS for genotypic prediction.

The results showed that WGS predicted rifampicin resistance with 100% sensitivity and 99.6% specificity, isoniazid resistance with 100% sensitivity and 100% specificity, ethambutol resistance with 100% sensitivity and 98.5% specificity, and pyrazinamide resistance with 93.8% sensitivity and 100% specificity. The line probe assay performed more poorly, with sensitivity of 83.3% for rifampicin and 85.7% for isoniazid resistance.

Perhaps more importantly, the study identified five isolates classified as susceptible by phenotypic DST that carried resistance mutations detected by WGS. One isolate had an rpoB mutation associated with rifampicin resistance, and four isolates had embB mutations associated with ethambutol resistance. These discordant cases suggest that WGS can correct phenotypic DST results, potentially identifying resistance that conventional testing misses.

The authors concluded that WGS, used under appropriate quality-control conditions, has good performance for predicting the resistance profile for the four first-line drugs and can correct phenotypic DST results. They highlighted the need for future guidelines recommending WGS as the initial tool in routine practice in areas where the prevalences of tuberculosis and drug-resistant MTB are low.

WGS and Treatment Outcomes in Multidrug-Resistant Tuberculosis

Beyond diagnostic accuracy, evidence links WGS-based susceptibility testing to improved clinical outcomes. A cohort study conducted in the Kyrgyz Republic, a multidrug-resistant tuberculosis hotspot in Central Asia, investigated whether the number of drugs tested susceptible by WGS and used in treatment regimens affected treatment response. The study included 227 participants with rifampicin-resistant or multidrug-resistant tuberculosis.

The researchers defined treatment response as a match between the drugs to which the isolate was susceptible and the drugs actually used in the regimen. They performed logistic regression to assess the effect of having at least four susceptible drugs in a regimen at baseline and at month two. The age- and sex-adjusted analysis showed an association between a regimen with at least four WGS matches at baseline and positive treatment response, with an adjusted odds ratio of 2.10 and a 95% confidence interval of 1.00 to 4.41. Notably, no association was found when conventional DST was used to define matches.

This finding carries significant clinical weight. It suggests that WGS-based susceptibility testing identifies drugs that are genuinely likely to be effective, and that constructing regimens based on WGS results improves the probability of treatment success. The study confirms that inclusion of four efficacious anti-tuberculosis drugs in a multidrug-resistant tuberculosis regimen increases the chances of a positive treatment response, and that susceptibility of at least four drugs as determined by WGS predicts a positive treatment response.

What These Studies Mean for Laboratory Decision-Making

Taken together, these studies support several practical conclusions for laboratories considering WGS adoption. First, WGS performs at least as well as line probe assays for detecting resistance to first-line drugs, and it outperforms them for rifampicin and isoniazid. Second, WGS can identify resistance mutations that phenotypic testing misses, suggesting that it may serve as a corrective tool instead of merely a confirmatory one. Third, WGS-based susceptibility predictions correlate with treatment outcomes, providing evidence that the genotypic approach has clinical validity beyond its technical accuracy.

However, the evidence also reveals important limitations. The sensitivity for pyrazinamide resistance was lower than for other first-line drugs, reflecting the complexity of pncA mutation interpretation. Discordance between WGS and phenotype occurs in both directions, and laboratories must have protocols for resolving these discrepancies. The studies were conducted in specific epidemiological contexts, and performance may differ in settings with different circulating strains and resistance mechanisms.

Bioinformatics Pipelines for Resistance Prediction

From Sequencing Reads to Resistance Calls

The journey from raw sequencing data to a clinically actionable susceptibility prediction requires a bioinformatics pipeline. This pipeline performs several essential functions: quality assessment of raw reads, alignment to a reference genome, variant calling, and annotation of mutations against resistance catalogues. Each step introduces potential sources of error, and laboratories must validate their pipelines before using them for clinical decisions.

The first step involves assessing the quality of the sequencing data. Metrics such as read length, base quality scores, and coverage depth determine whether the data are suitable for downstream analysis. Low-quality reads can introduce spurious variants, while insufficient coverage can cause true variants to be missed. Laboratories must establish minimum thresholds for these metrics based on their sequencing platform and validation data.

The second step aligns the sequencing reads to a reference genome. For Mycobacterium tuberculosis, the standard reference is the H37Rv strain, which has been fully sequenced and annotated. Alignment algorithms must account for the high GC content of the mycobacterial genome and the presence of repetitive elements that can complicate read placement.

The third step identifies variants by comparing the aligned reads to the reference genome. This step requires careful filtering to distinguish true biological variants from sequencing errors. Variant callers apply statistical models to assess the probability that each observed difference represents a genuine mutation instead of an artifact.

The fourth step annotates the identified variants and compares them against curated catalogues of resistance-associated mutations. Tools such as TBProfiler, which was used in the Papua study, automate this process by maintaining databases of known resistance mutations and generating predicted susceptibility profiles. The accuracy of this step depends entirely on the completeness and accuracy of the underlying mutation catalogue.

The Challenge of Mutation Catalogue Completeness

The reliability of WGS-based resistance prediction depends on the completeness of the mutation catalogue used for interpretation. For well-characterized drugs such as rifampicin and isoniazid, the catalogue is relatively complete, and most resistance mechanisms have been identified. For other drugs, particularly newer and repurposed agents, the genetic basis of resistance is less fully understood, and prediction accuracy may be lower.

This limitation has practical consequences. A WGS result that predicts susceptibility does not guarantee that the isolate is truly susceptible, because the isolate may carry a resistance mechanism that has not yet been characterized. Conversely, a WGS result that predicts resistance is generally reliable, because the presence of a known resistance mutation provides strong evidence of phenotypic resistance.

Laboratories must communicate this distinction to clinicians. A WGS-based susceptibility prediction should be interpreted as a probabilistic statement based on current knowledge, not as an absolute determination. When treatment decisions carry high stakes, discordant or unexpected results should prompt confirmatory testing.

Quality Control for Bioinformatics Analysis

Bioinformatics quality control extends beyond the sequencing run itself. Laboratories must validate their pipelines against reference datasets with known resistance profiles, establish thresholds for variant quality scores and coverage, and participate in external quality assessment programs when available. The published WGS workflow emphasizes that run evaluations must be customized by each laboratory, reflecting the need for local validation instead of reliance on generic protocols.

Version control is another critical consideration. Mutation catalogues and bioinformatics tools are updated regularly as new resistance mechanisms are discovered. Laboratories must document which versions of which tools were used for each analysis, because a prediction made with one version of a catalogue may differ from a prediction made with a later version. This documentation is essential for interpreting results retrospectively and for maintaining consistency in longitudinal studies.

Practical Implementation of WGS in a Mycobacteriology Laboratory

Step 1: Define the Clinical Question and Scope

Before investing in WGS infrastructure, a laboratory must define the clinical question it intends to answer. Will WGS replace phenotypic DST for first-line drugs, or will it serve as an adjunct for specific scenarios such as multidrug-resistant tuberculosis? Will the laboratory sequence all isolates or only those with suspected resistance? The answers to these questions determine the required throughput, turnaround time, and bioinformatics capacity.

The French routine laboratory study provides a model for one approach: sequencing all MTB isolates and using WGS as the initial tool for first-line drug susceptibility prediction. This approach maximizes the clinical benefit of rapid results but requires sufficient sequencing capacity to handle the full sample volume. Laboratories with lower throughput might adopt a targeted approach, sequencing only isolates from patients with risk factors for drug resistance or from cases where phenotypic DST has already indicated resistance.

Step 2: Assess Infrastructure Requirements

WGS requires substantial infrastructure investment. The sequencing instrument itself represents a significant capital cost, and the laboratory must also acquire DNA extraction equipment, quantification instruments, and computational resources for data analysis. The published workflow notes that reagent costs can be reduced through protocol optimization, but the initial investment remains substantial.

Personnel requirements are equally important. WGS implementation requires staff with molecular biology skills for library preparation, bioinformatics skills for data analysis, and clinical microbiology skills for result interpretation. The American Society for Microbiology emphasizes the importance of professional education and workforce development in microbiology, and laboratories must invest in training before implementing WGS.

Biosafety considerations also apply. The Biosafety in Microbiological and Biomedical Laboratories guidance from CDC and NIH provides the framework for risk assessment and containment when working with Mycobacterium tuberculosis cultures. DNA extraction from mycobacterial cultures must be performed under appropriate containment conditions, and the laboratory must have decontamination procedures for spills and waste.

Step 3: Validate the Complete Workflow

Validation is the most critical step in WGS implementation. The laboratory must demonstrate that its complete workflow, from DNA extraction through bioinformatics analysis, produces accurate and reproducible results. Validation should include a panel of well-characterized isolates with known resistance profiles, including isolates resistant to each drug the laboratory intends to predict.

The validation panel should include isolates from the American Type Culture Collection or other authenticated culture collections to ensure that reference strains are genuine and well characterized. The laboratory should also include clinical isolates with diverse resistance mechanisms to test the pipeline's ability to detect different mutations.

Validation metrics should include sensitivity and specificity for each drug, positive and negative predictive values, and reproducibility across replicate runs. The laboratory should also assess the limit of detection for mixed infections or heteroresistance, where a sample contains both susceptible and resistant populations.

Step 4: Establish Quality Control Procedures

Ongoing quality control is essential for maintaining the accuracy of WGS-based predictions. The laboratory should include control samples in each sequencing run, monitor sequencing metrics such as coverage and error rates, and participate in external quality assessment programs when available.

The published WGS workflow emphasizes that quality controls, instrument parameters, and run evaluations must be customized by each laboratory. This customization should be documented in standard operating procedures that specify acceptable ranges for each metric and actions to take when metrics fall outside those ranges.

Quality control also extends to the bioinformatics pipeline. The laboratory should monitor the performance of its variant calling and resistance prediction tools, tracking any changes in results that might indicate software updates or database changes. Regular revalidation should be performed whenever the pipeline is updated.

Step 5: Develop Reporting and Interpretation Protocols

The laboratory must develop clear protocols for reporting WGS-based susceptibility results to clinicians. Reports should include the predicted susceptibility profile for each drug, the specific mutations detected, and any limitations or caveats relevant to interpretation. The report should distinguish between drugs for which the genetic basis of resistance is well characterized and drugs for which prediction is less reliable.

The laboratory should also establish protocols for handling discordant results. When WGS predicts resistance but phenotypic DST indicates susceptibility, or vice versa, the laboratory should repeat both tests and consider additional testing such as minimum inhibitory concentration determination. The French study identified five isolates where WGS detected resistance mutations despite phenotypic susceptibility, and the laboratory must have a process for investigating such cases.

Records and Measurements for WGS-Based Susceptibility Testing

Essential Data to Document

Laboratories implementing WGS for mycobacterial susceptibility testing must maintain comprehensive records that support both clinical decision-making and quality assurance. The following data elements should be documented for each isolate:

Quality Metrics to Monitor

The laboratory should track quality metrics at each stage of the WGS workflow. DNA extraction quality can be assessed through spectrophotometric ratios and fluorometric quantification. Library quality can be assessed through fragment size distribution and concentration measurements. Sequencing quality can be assessed through read depth, coverage uniformity, and error rates.

The published WGS workflow emphasizes that instrument parameters and run evaluations must be customized by each laboratory. This customization should include establishing thresholds for acceptable performance and documenting actions taken when thresholds are not met. The laboratory should review these metrics regularly to identify trends that might indicate emerging problems with reagents, instruments, or procedures.

Longitudinal Performance Tracking

Beyond individual run metrics, the laboratory should track its overall performance over time. This tracking should include the distribution of resistance predictions across drugs, the frequency of discordant results with phenotypic testing, and the turnaround time from specimen receipt to result reporting. These data support continuous quality improvement and provide evidence of the laboratory's performance for accreditation and regulatory purposes.

The laboratory should also track the clinical impact of its WGS service. The Kyrgyz Republic study demonstrated that WGS-based susceptibility testing was associated with improved treatment outcomes, and the laboratory should seek feedback from clinicians about how WGS results influenced treatment decisions. This feedback loop helps the laboratory refine its reporting and interpretation protocols to better serve clinical needs.

Common Failure Patterns and Troubleshooting

DNA Quality Failures

Poor DNA quality is a common cause of WGS failure. Mycobacterial DNA can be difficult to extract because of the organism's thick, lipid-rich cell wall, and the extraction method must be optimized for this organism. Inadequate cell lysis results in low DNA yield, while excessive shearing produces fragments that are too short for library preparation.

The laboratory should monitor DNA quantity and quality at the extraction step and establish minimum thresholds for proceeding to library preparation. If DNA quality is inadequate, the laboratory should repeat the extraction, potentially using a different method or additional mechanical disruption. The published WGS workflow notes that protocol details often need to be developed and customized by each laboratory, and DNA extraction is a prime example of a step that may require optimization.

Library Preparation Failures

Library preparation failures can manifest as low library yield, incorrect fragment size distribution, or adapter contamination. These failures can result from reagent degradation, incorrect reagent volumes, or problems with the quantification step. The published workflow emphasizes that quality controls must be established for each step, and the laboratory should have clear criteria for accepting or rejecting libraries.

When library preparation fails, the laboratory should systematically troubleshoot each step of the process. This troubleshooting should include checking reagent expiration dates, verifying instrument calibration, and reviewing the technician's technique. The laboratory should document the troubleshooting process and any protocol modifications that resolve the problem.

Sequencing Run Failures

Sequencing run failures can result from instrument problems, reagent issues, or sample-related factors. Common symptoms include low cluster density, poor base quality scores, or run abortion. The laboratory should monitor run metrics in real time and have contingency plans for rerunning failed samples.

The published WGS workflow emphasizes that run evaluations must be customized by each laboratory. This customization should include establishing thresholds for acceptable run metrics and documenting actions taken when runs fail. The laboratory should also maintain relationships with instrument manufacturers for technical support and troubleshooting guidance.

Bioinformatics Pipeline Errors

Bioinformatics errors can produce incorrect resistance predictions even when sequencing quality is high. These errors can result from software bugs, incorrect parameter settings, or outdated mutation catalogues. The laboratory should validate its pipeline against reference datasets and participate in external quality assessment programs to detect systematic errors.

When a resistance prediction seems inconsistent with phenotypic results or clinical expectations, the laboratory should review the raw sequencing data and the bioinformatics analysis. This review should include checking read quality, coverage at the relevant genomic positions, and the variant calling parameters. The laboratory should document any pipeline errors and implement corrective actions to prevent recurrence.

Limitations of WGS for Mycobacterial Susceptibility Testing

Incomplete Knowledge of Resistance Mechanisms

The most fundamental limitation of WGS-based susceptibility testing is that the genetic basis of resistance is not fully characterized for all drugs. For first-line drugs such as rifampicin and isoniazid, the major resistance mechanisms are well understood, and prediction accuracy is high. For other drugs, particularly newer agents and those with complex resistance mechanisms, the catalogue of known resistance mutations is incomplete.

This limitation means that a WGS prediction of susceptibility does not guarantee phenotypic susceptibility. The isolate may carry a resistance mechanism that has not yet been identified, or the resistance may result from a mechanism that is not detectable by sequencing, such as efflux pump overexpression or changes in drug permeability. Laboratories must communicate this uncertainty to clinicians and recommend phenotypic confirmation when treatment decisions are critical.

Discordance Between Genotype and Phenotype

Discordance between WGS predictions and phenotypic DST results occurs in both directions. The French study identified isolates where WGS detected resistance mutations despite phenotypic susceptibility, suggesting that phenotypic testing can miss resistance. Conversely, phenotypic resistance can occur without detectable resistance mutations, suggesting that some resistance mechanisms remain unidentified.

The clinical significance of discordant results depends on the specific drug and the clinical context. For rifampicin, resistance is almost always associated with rpoB mutations, and a WGS prediction of resistance is highly reliable. For other drugs, the correlation between genotype and phenotype is less strong, and discordant results require careful interpretation.

The laboratory must have protocols for resolving discordant results. These protocols should include repeat testing, additional molecular testing, and clinical correlation. The laboratory should also document discordant cases and review them periodically to identify patterns that might indicate problems with either the WGS or phenotypic testing methods.

Technical Limitations of Sequencing

Technical limitations of sequencing can affect the accuracy of resistance predictions. Coverage depth must be sufficient to detect mutations with confidence, particularly in mixed infections where resistant and susceptible populations coexist. The GC-rich nature of the mycobacterial genome can cause sequencing biases that reduce coverage in certain regions.

The laboratory must establish minimum coverage thresholds for reliable variant calling and monitor coverage across the genome to identify regions with inadequate depth. When coverage is insufficient, the laboratory should repeat the sequencing or use targeted approaches to interrogate specific resistance-associated genes.

Cost and Infrastructure Barriers

The cost of WGS equipment, reagents, and bioinformatics infrastructure remains a barrier for many laboratories. The published workflow describes an adapted protocol optimized for reduced reagent costs, acknowledging that cost is a significant consideration for WGS implementation. Laboratories must conduct a cost-benefit analysis that considers their sample volume, the prevalence of drug resistance in their population, and the clinical value of faster results.

The cost of WGS must be weighed against the costs of prolonged ineffective treatment, which include additional healthcare utilization, lost productivity, and the public health consequences of continued transmission. In settings with high rates of drug-resistant tuberculosis, the clinical and economic benefits of rapid WGS-based susceptibility testing may justify the infrastructure investment.

Biosafety and Regulatory Considerations

Containment Requirements for Mycobacterial Work

Laboratories working with Mycobacterium tuberculosis must follow established biosafety guidelines. The Biosafety in Microbiological and Biomedical Laboratories guidance from CDC and NIH provides the framework for risk assessment and containment. Mycobacterium tuberculosis requires Biosafety Level 3 practices for manipulation of cultures, including appropriate respiratory protection and containment equipment.

WGS workflows involve multiple steps that require different levels of containment. DNA extraction from cultures must be performed under BSL-3 conditions because the initial steps involve handling viable organisms. Once DNA has been extracted and the organism has been inactivated, subsequent steps such as library preparation and sequencing can be performed under lower containment conditions.

The laboratory must have clear protocols for the transition from BSL-3 to lower containment conditions, including verification that the organism has been inactivated. The published WGS workflow should include specific instructions for this transition, and the laboratory must document compliance with biosafety requirements.

Quality Standards and Accreditation

Laboratories implementing WGS for clinical use must operate under appropriate quality standards. The Clinical and Laboratory Standards Institute provides standards for antimicrobial susceptibility testing, and these standards provide a framework for quality assurance even when the testing method is genotypic instead of phenotypic. The laboratory should align its WGS validation and quality control procedures with CLSI expectations.

Accreditation bodies may have specific requirements for WGS-based testing. The laboratory should consult its accreditation body early in the implementation process to understand these requirements and ensure that its validation and quality control procedures will satisfy them. The American Society for Microbiology provides professional education and resources that can support laboratory staff in meeting these requirements.

Regulatory Status of WGS-Based Susceptibility Testing

The regulatory status of WGS-based susceptibility testing varies by jurisdiction and by the intended use of the results. Laboratories must understand the regulatory requirements that apply to their specific context, including whether WGS-based predictions require confirmation by phenotypic testing for clinical decision-making.

The published studies describe WGS as a tool that can correct phenotypic DST results, suggesting that it has clinical utility beyond simple confirmation. However, the regulatory framework for such use may require specific validation and approval. Laboratories should consult with their regulatory authorities and professional organizations to understand the applicable requirements.

Professional Escalation Criteria

When to Seek Additional Expertise

Laboratory professionals should recognize situations that require escalation to additional expertise. These situations include persistent discordance between WGS and phenotypic results, unusual resistance patterns, and cases where WGS results have major implications for patient management or public health.

The laboratory should have established relationships with reference laboratories and subject matter experts who can provide consultation for challenging cases. The American Type Culture Collection provides authenticated reference strains that can be used for validation and troubleshooting, and the National Center for Biotechnology Information maintains the Pathogen Detection database that supports genomic epidemiology and resistance surveillance.

When to Repeat Testing

Repeat testing is appropriate when sequencing quality is inadequate, when coverage is insufficient at resistance-associated loci, or when discordant results cannot be resolved through initial investigation. The laboratory should have clear criteria for repeat testing and should document the reasons for repeating any sample.

Repeat testing is also appropriate when the clinical context suggests that the WGS result may be incorrect. For example, if a patient has documented exposure to a drug-resistant case but the WGS result predicts susceptibility, the laboratory should consider repeat testing and additional investigation.

When to Involve Public Health Authorities

Certain WGS results have public health implications that require notification of public health authorities. These include detection of extensively drug-resistant tuberculosis, identification of resistance to drugs used in standard regimens, and detection of resistance patterns that suggest transmission of drug-resistant strains.

The World Health Organization emphasizes the importance of surveillance for antimicrobial resistance, and WGS data can contribute to this surveillance by providing detailed information about resistance mechanisms and strain relationships. The laboratory should have protocols for reporting WGS results to public health authorities in accordance with local requirements.

Frequently Asked Questions

How does whole-genome sequencing predict drug susceptibility in Mycobacterium tuberculosis?

WGS reads the complete genetic code of an isolate and searches for mutations in genes known to confer drug resistance. Bioinformatics tools compare detected mutations against curated catalogues of resistance-associated variants and generate a predicted susceptibility profile for each drug.

Is WGS more accurate than phenotypic drug susceptibility testing?

Published evidence shows WGS achieves sensitivity of 100% for rifampicin and isoniazid resistance and 93.8% for pyrazinamide resistance when compared against phenotypic testing as the reference. WGS can also detect resistance mutations that phenotypic testing misses, suggesting it can correct false-susceptible results.

How long does WGS-based susceptibility testing take compared to phenotypic methods?

WGS can produce results in days after DNA extraction and sequencing, while phenotypic DST requires weeks because of the slow growth of Mycobacterium tuberculosis. This faster turnaround enables earlier regimen adjustment for patients with drug-resistant disease.

What are the main limitations of WGS for mycobacterial susceptibility testing?

The main limitations are incomplete knowledge of resistance mechanisms for some drugs, discordance between genotypic predictions and phenotypic results, technical requirements for adequate sequencing coverage, and the cost of equipment and bioinformatics infrastructure.

Can WGS replace phenotypic drug susceptibility testing entirely?

Current evidence supports WGS as an initial tool for first-line drug susceptibility prediction, but phenotypic testing remains important for confirming results and for drugs where the genetic basis of resistance is not fully characterized. The French routine laboratory study recommends WGS as the initial tool in low-prevalence settings.

What bioinformatics tools are used for WGS-based resistance prediction?

Tools such as TBProfiler automate the process of comparing detected mutations against resistance catalogues. The Papua study used TBProfiler for genome analysis, and the laboratory must validate its chosen pipeline against reference datasets with known resistance profiles.

How should discordant WGS and phenotypic results be handled?

Discordant results should trigger repeat testing, additional molecular investigation, and clinical correlation. The laboratory should document discordant cases and review them periodically to identify patterns that might indicate problems with either testing method.

What biosafety level is required for WGS workflows with Mycobacterium tuberculosis?

DNA extraction from cultures requires Biosafety Level 3 practices because it involves handling viable organisms. Once DNA has been extracted and the organism inactivated, subsequent steps such as library preparation and sequencing can be performed under lower containment conditions.

Related Bacteriology Guides

• Biological Foundation Models for Antimicrobial Peptide Discovery in Veterinary Pathogens
• Aeromonas hydrophila in Aquaculture: Virulence Factors and Antimicrobial Resistance
• Common Bacterial Pathogens in Chickens: Identification and Antimicrobial Management
• Swine Bacterial Diseases: Colibacillosis, UTI, and Antimicrobial Resistance
• Avian Colibacillosis: Pathogenesis, Diagnosis, and Antimicrobial Resistance Patterns in Poultry

References and Further Reading

• Biosafety in Microbiological and Biomedical Laboratories. CDC and NIH.
• Microbiology Resources. American Society for Microbiology.
• Antimicrobial Resistance. World Health Organization.
• Bacteriological Analytical Manual. U.S. Food and Drug Administration.
• Antimicrobial Susceptibility Testing. Clinical and Laboratory Standards Institute.
• Culture Collections. American Type Culture Collection.
• NCBI Pathogen Detection. National Center for Biotechnology Information.
• Use of Whole Genome Sequencing for Mycobacterium tuberculosis Complex Antimicrobial Susceptibility Testing.. Methods in molecular biology (Clifton, N.J.), 2024.
• Whole-genome sequencing drug susceptibility testing is associated with positive MDR-TB treatment response.. The international journal of tuberculosis and lung disease : the official journal of the International Union against Tuberculosis and Lung Disease, 2024.
• Whole-genome sequencing in drug susceptibility testing of Mycobacterium tuberculosis in routine practice in Lyon, France.. International journal of antimicrobial agents, 2020.
• The whole-genome sequencing in predicting Mycobacterium tuberculosis drug susceptibility and resistance in Papua, Indonesia.. BMC genomics, 2021.

This article is educational and does not replace validated laboratory procedures, clinical diagnosis, institutional biosafety review, or public-health guidance.