Choosing the Right Sequencing Platform for AMR Surveillance

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-08-27

Quick Answer

• Choose Illumina short-read sequencing for routine AMR surveillance when you need high accuracy and low cost per sample at scale.
• Select Oxford Nanopore long-read sequencing when you must resolve plasmid structures and genomic context around resistance genes.
• Match platform choice to your surveillance scale, budget, and bioinformatics capacity because each platform has distinct tradeoffs in read length, accuracy, and cost.

Understanding Sequencing Platforms for AMR Surveillance

Antimicrobial resistance surveillance depends on whole-genome sequencing to identify resistance genes, track their spread, and understand the genetic mechanisms that make bacteria resistant to antibiotics. The World Health Organization identifies antimicrobial resistance as a global public health threat and emphasizes that surveillance systems are essential for understanding resistance patterns and guiding stewardship efforts. Laboratories planning genomic surveillance must select a sequencing platform that balances cost, throughput, and accuracy for resistance gene detection.

The three main sequencing platforms used in bacterial genomics are Illumina short-read sequencing, Oxford Nanopore long-read sequencing, and PacBio long-read sequencing. Each platform generates different types of data that affect how well you can detect resistance genes, assemble genomes, and interpret results. The choice of platform influences your laboratory workflow, data analysis requirements, and the types of questions you can answer about resistant bacteria.

This article compares the three main platforms in terms of read length, accuracy, cost per sample, and suitability for different surveillance scales. It provides practical guidance for laboratories planning AMR surveillance programs, including how to assess your needs, what records to keep, and when to escalate technical problems.

At a Glance

The table below summarizes the key characteristics of the three main sequencing platforms for AMR surveillance. These comparisons reflect general platform capabilities and should be verified with current manufacturer specifications before making purchasing decisions.

The choice between these platforms depends on your surveillance objectives. If your goal is to detect resistance genes across many samples at the lowest cost per sample, short-read sequencing is often the most practical option. If you need to understand how resistance genes are carried and spread, long-read sequencing provides the genomic context that short reads cannot resolve.

Core Principles of Sequencing for AMR Surveillance

What Whole-Genome Sequencing Reveals About Resistance

Whole-genome sequencing provides a complete picture of the bacterial genome, allowing you to identify the specific genes that confer resistance to antibiotics. Unlike traditional antimicrobial susceptibility testing, which measures the phenotypic response of bacteria to drugs, genomic sequencing reveals the genetic basis of resistance. This information helps you understand which resistance mechanisms are present, how they are acquired, and how they may spread between bacteria.

The World Health Organization emphasizes that surveillance of antimicrobial resistance requires reliable laboratory data and standardized methods. Whole-genome sequencing is increasingly used as a tool for AMR surveillance because it provides detailed information about resistance genes and their genetic context. However, sequencing data must be interpreted in the context of phenotypic susceptibility testing to confirm that the genes you identify actually produce resistance in the organism.

How Read Length Affects Resistance Gene Detection

Read length is one of the most important factors in sequencing platform selection. Short reads of 150 to 300 base pairs are sufficient to identify most resistance genes because these genes are typically several hundred to several thousand base pairs long. When you sequence a bacterial genome with short reads, you can align the reads to a reference database of known resistance genes and identify which genes are present.

However, short reads have limitations when you need to understand the genomic context of resistance genes. Resistance genes are often located on mobile genetic elements such as plasmids, transposons, and integrons. These elements can move between bacteria and spread resistance across species. Short reads may not be able to assemble these elements completely because they contain repetitive sequences that are longer than the read length.

Long-read sequencing platforms generate reads that span these repetitive regions, allowing you to assemble complete plasmids and other mobile genetic elements. This information is valuable for understanding how resistance spreads and for tracking outbreaks. The tradeoff is that long-read sequencing is generally more expensive per sample and may require more sophisticated bioinformatics analysis.

Core Principles of Platform Selection

Accuracy and Error Profiles

The accuracy of sequencing data is critical for AMR surveillance because you need to identify resistance genes with confidence. Short-read platforms such as Illumina have high per-base accuracy, typically exceeding 99.9 percent. This means that the probability of a sequencing error at any given position is very low, and you can confidently identify single nucleotide variants and resistance genes.

Long-read platforms have different error profiles. Oxford Nanopore sequencing has improved significantly in recent years, but raw read accuracy is lower than short-read platforms. The error rate depends on the specific chemistry and basecalling software used. PacBio sequencing with circular consensus sequencing can achieve accuracy comparable to short-read platforms, but this approach is more expensive per sample.

For AMR surveillance, the accuracy of resistance gene detection depends on both the sequencing platform and the bioinformatics pipeline. You should validate your pipeline using reference strains with known resistance genes to ensure that your analysis correctly identifies the genes present. The American Type Culture Collection provides authenticated biological materials that can be used for quality control and validation of sequencing workflows.

Throughput and Scalability

Throughput refers to the number of samples you can sequence in a given time period. This is an important consideration for AMR surveillance because you may need to sequence hundreds or thousands of samples to understand resistance patterns in a population.

Illumina platforms offer high throughput, allowing you to sequence many samples in parallel. The cost per sample decreases as you increase the number of samples sequenced in a single run. This makes short-read sequencing well suited for large-scale surveillance programs where you need to process many samples efficiently.

Oxford Nanopore platforms offer flexible throughput, with instruments that can run from a single sample to many samples depending on the flow cell used. The ability to sequence in real time can be useful for surveillance programs that need rapid results. However, the cost per sample may be higher than Illumina for large-scale projects.

PacBio platforms typically have lower throughput than Illumina but produce longer reads. This makes them well suited for projects that require complete genome assemblies or detailed analysis of genomic structure. The higher cost per sample means that PacBio is often used for smaller numbers of samples that require the most detailed analysis.

Cost Analysis for AMR Surveillance Programs

Instrument and Consumable Costs

The initial investment in sequencing instruments varies significantly between platforms. Illumina instruments range from benchtop models suitable for small laboratories to production-scale instruments for large sequencing centers. The capital cost of the instrument is an important consideration, but the ongoing cost of consumables and reagents is often the larger factor in the total cost of a surveillance program.

Oxford Nanopore instruments have a lower initial cost than many Illumina platforms, and the consumable costs are based on the flow cells and sequencing kits used. The cost per sample depends on how many samples you can multiplex in a single flow cell. For small numbers of samples, the cost per sample may be higher than Illumina because you are paying for the flow cell regardless of how many samples you sequence.

PacBio instruments have a higher capital cost, and the consumable costs are also higher than other platforms. The cost per sample is typically the highest of the three platforms, but the data quality and read length may justify the cost for certain applications.

Total Cost of Ownership

When comparing platforms, you should consider the total cost of ownership, which includes the instrument, consumables, maintenance, and the cost of bioinformatics infrastructure. The bioinformatics cost is often overlooked but can be significant. You need computing resources to process sequencing data, storage for the data, and personnel with the skills to run the analysis.

The American Society for Microbiology emphasizes the importance of laboratory workforce development and professional education. The bioinformatics skills required for genomic surveillance are specialized, and you may need to invest in training or hire personnel with these skills. The cost of bioinformatics should be included in your budget when comparing platforms.

Cost per Sample at Different Scales

The cost per sample varies with the scale of your surveillance program. For small numbers of samples, the cost per sample is higher because you are paying for the instrument and consumables regardless of the number of samples. As you increase the number of samples, the cost per sample decreases because you can spread the fixed costs across more samples.

For large-scale surveillance programs, Illumina short-read sequencing typically offers the lowest cost per sample. This is because the platform supports high levels of multiplexing, allowing you to sequence many samples in a single run. The cost per sample can be reduced further by using smaller sequencing kits or by pooling samples.

For smaller surveillance programs, the cost per sample may be less important than the flexibility and speed of the platform. Oxford Nanopore sequencing can be a cost-effective option for small numbers of samples because you can sequence a single sample or a small number of samples without the need for a large batch.

Read Length and Genomic Resolution

Short Reads and Resistance Gene Detection

Short-read sequencing is well suited for detecting resistance genes because most resistance genes are shorter than the read length. You can align short reads to a database of known resistance genes and identify which genes are present in your sample. This approach is reliable for detecting genes that are already known and included in the database.

Short reads are also useful for detecting single nucleotide variants in resistance genes. These variants can change the function of a gene and affect the level of resistance. By comparing the sequence of a resistance gene to a reference sequence, you can identify variants that may be associated with resistance.

However, short reads have limitations when you need to determine the genomic context of a resistance gene. If a resistance gene is located on a plasmid, you may not be able to determine which plasmid it is on or how the plasmid is structured. This information is important for understanding how resistance spreads between bacteria.

Long Reads and Genomic Context

Long-read sequencing provides the genomic context that short reads cannot. With reads of 10,000 base pairs or more, you can assemble complete plasmids and other mobile genetic elements. This allows you to determine which plasmid carries a resistance gene, the structure of the plasmid, and the other genes that are present on the same element.

This information is valuable for AMR surveillance because it helps you understand how resistance spreads. If you find the same resistance gene on the same plasmid in different bacterial species, this suggests that the plasmid is spreading between species. This information can inform infection control and stewardship efforts.

Long-read sequencing also allows you to assemble complete bacterial genomes, which can be useful for understanding the genetic basis of resistance. A complete genome assembly provides a reference for the organism and allows you to identify all resistance genes and their genomic context.

Bioinformatics Requirements for AMR Surveillance

Data Analysis Pipeline

The bioinformatics pipeline for AMR surveillance includes several steps: quality control, read trimming, assembly, and resistance gene detection. The specific tools you use depend on the sequencing platform and the type of data you generate.

For short-read data, the pipeline typically includes quality control to remove low-quality reads, trimming to remove adapter sequences, and assembly to reconstruct the genome. The assembled genome is then compared to a database of resistance genes to identify which genes are present.

For long-read data, the pipeline may include additional steps to correct errors in the reads and to assemble the genome. The assembly of long-read data can produce complete genomes, but the error rate of the raw reads must be corrected to produce an accurate assembly.

Bioinformatics Skills and Infrastructure

The bioinformatics skills required for AMR surveillance are specialized. You need to be able to run the analysis pipeline, interpret the results, and troubleshoot problems. The American Society for Microbiology provides professional education and resources for laboratory professionals, which can help you develop the skills you need.

The infrastructure required for bioinformatics analysis includes computing resources, storage, and software. The amount of computing power you need depends on the number of samples you are processing and the complexity of the analysis. For large-scale surveillance, you may need access to a high-performance computing cluster.

Data Storage and Management

The data generated by sequencing is large. A single bacterial genome sequenced with short reads generates hundreds of megabytes of data. A long-read sequencing run can generate even more data. You need to have a plan for storing and managing this data, including backup and security.

The data management plan should include a system for organizing the data by sample, tracking the analysis results, and ensuring that the data is accessible for future analysis. The data should be stored in a format that is compatible with the bioinformatics tools you are using.

Practical Workflow for Platform Selection

Step 1: Define Your Surveillance Objectives

Before selecting a platform, you need to define your surveillance objectives. What questions are you trying to answer? Are you trying to detect resistance genes in a large number of samples, or are you trying to understand the genomic context of resistance in a smaller number of samples? The answers to these questions will guide your platform selection.

If your objective is to detect resistance genes in a large number of samples, short-read sequencing is likely the most appropriate choice. If your objective is to understand the genomic context of resistance, long-read sequencing may be necessary.

Step 2: Assess Your Budget and Resources

Assess your budget for the sequencing platform, including the instrument, consumables, and bioinformatics infrastructure. Consider the total cost of ownership, beyond the initial purchase price. The cost per sample is an important factor, but the total cost of the program is what matters.

Step 3: Evaluate Your Bioinformatics Capacity

Evaluate your bioinformatics capacity, including the skills of your personnel and the computing resources available. The bioinformatics requirements for long-read data are different from those for short-read data. Make sure you have the skills and resources to analyze the data you generate.

Step 4: Consider the Throughput Requirements

Consider the throughput requirements of your surveillance program. How many samples do you need to sequence per week or per month? The throughput of the platform must match your requirements. If you need to sequence many samples, a high-throughput platform is necessary.

Step 5: Validate the Platform with Reference Strains

Before you commit to a platform, validate it with reference strains that have known resistance genes. The American Type Culture Collection provides authenticated biological materials that can be used for this purpose. The validation should confirm that the platform can detect the resistance genes you expect to find.

Step 6: Make the Decision and Plan the Implementation

Based on your objectives, budget, and capacity, make the decision and plan the implementation. The implementation plan should include the purchase of the instrument, the training of personnel, and the development of the bioinformatics pipeline.

Records and Measurements for Platform Evaluation

Key Metrics to Track

When evaluating a sequencing platform for AMR surveillance, track the following metrics:

• Read length: The average and maximum read length for each run.
• Read accuracy: The per-base accuracy of the sequencing data.
• Cost per sample: The total cost of sequencing, including consumables and labor, divided by the number of samples.
• Throughput: The number of samples sequenced per run or per week.
• Turnaround time: The time from sample receipt to the final result.
• Resistance gene detection rate: The number of resistance genes detected in a set of reference samples.

Keeping Records

Keep records of the sequencing runs, including the platform, the chemistry, the quality metrics, and the results. These records are important for quality control and for troubleshooting problems. The records should be stored in a way that is accessible for future analysis.

Quality Control

Quality control is essential for AMR surveillance. The quality of the sequencing data affects the accuracy of the resistance gene detection. Use reference strains to validate the platform and the pipeline, and monitor the quality metrics on each run.

The Clinical and Laboratory Standards Institute provides standards for antimicrobial susceptibility testing, which can be used to validate the phenotypic results. The sequencing results should be compared to the phenotypic results to confirm that the resistance genes you detect are associated with resistance.

Common Failure Patterns and How to Avoid Them

Low-Quality Sequencing Data

Low-quality sequencing data can result from problems with the sequencing chemistry, the sample preparation, or the instrument. The quality of the data can be assessed by the quality metrics, such as the quality score and the read length. If the quality is low, the results may be unreliable.

To avoid low-quality data, follow the manufacturer's recommendations for sample preparation and sequencing. Use the quality controls to monitor the quality of the data and troubleshoot problems.

Incomplete Genome Assemblies

Incomplete genome assemblies can result from insufficient coverage or from repetitive regions that cannot be resolved. The assembly may be fragmented, and the resistance genes may not be fully assembled. This can affect the accuracy of the resistance gene detection.

To avoid incomplete assemblies, ensure that you have sufficient coverage of the genome. For short-read sequencing, this means sequencing to a sufficient depth. For long-read sequencing, the read length and the coverage determine the completeness of the assembly.

Misidentification of Resistance Genes

Misidentification of resistance genes can result from errors in the sequencing data or from the limitations of the detection database. The database may not contain all resistance genes, or the genes may be similar to other genes, leading to false positives or false negatives.

To avoid misidentification, use a comprehensive database of resistance genes and validate the results with phenotypic testing. The Clinical and Laboratory Standards Institute provides standards for antimicrobial susceptibility testing, which can be used to confirm the results.

High Cost per Sample

The cost per sample can be higher than expected if the throughput is lower than planned or if the consumable costs are higher than expected. To avoid high costs, plan the sequencing runs to maximize the number of samples per run and monitor the consumable costs.

Safety and Regulatory Context

Laboratory Biosafety

Sequencing of bacterial pathogens requires appropriate biosafety measures. The Centers for Disease Control and Prevention and the National Institutes of Health provide guidance on biosafety in microbiological and biomedical laboratories. The risk assessment should consider the organisms being sequenced and the procedures being performed.

The biosafety level of the laboratory depends on the organisms being handled. For most bacteria, biosafety level 2 is appropriate. The laboratory should have the appropriate containment and decontamination procedures in place.

Data Security and Privacy

The sequencing data may contain information that is sensitive, such as the identity of the samples and the resistance genes present. The data should be stored securely and access should be restricted to authorized personnel. The data management plan should include the security measures.

Regulatory Considerations

The use of sequencing for AMR surveillance may be subject to regulatory requirements. The requirements depend on the jurisdiction and the purpose of the surveillance. The laboratory should be aware of the regulatory requirements and ensure that the sequencing is performed in compliance with the requirements.

Professional Escalation Criteria

When to Escalate to a Specialist

If you encounter problems with the sequencing platform or the bioinformatics pipeline that you cannot resolve, escalate to a specialist. The specialist may be a bioinformatician, a sequencing specialist, or a microbiologist with expertise in AMR surveillance.

When to Escalate to a Public Health Authority

If you detect a resistance gene that is associated with a public health concern, you may need to escalate to a public health authority. The World Health Organization provides guidance on antimicrobial resistance surveillance and the public health response. The escalation criteria depend on the organism and the resistance gene.

When to Escalate to a Regulatory Authority

If you detect a resistance gene that is associated with a regulatory concern, you may need to escalate to a regulatory authority. The FDA provides guidance on laboratory methods for food microbiology, which may be relevant for foodborne pathogens.

A Decision Framework for Matching Platform Choice to Surveillance Scale

Selecting a sequencing platform for AMR surveillance requires more than comparing instrument specifications. Laboratories often struggle because they evaluate platforms in isolation instead of against the specific demands of their surveillance program. A practical decision framework helps you match platform capabilities to your operational reality, including sample volume, reporting deadlines, and the types of resistance questions you must answer.

The Three-Tier Surveillance Scale Model

Surveillance programs fall into three operational tiers based on sample volume and reporting frequency. Your tier determines which platform characteristics matter most.

Tier 1: Local or outbreak-focused surveillance. This tier processes fewer than 50 samples per month. The priority is rapid turnaround for outbreak investigations and the ability to sequence single samples without waiting to accumulate a batch. Oxford Nanopore sequencing fits this tier well because you can run one sample or a small set of samples on demand. The real-time data generation allows you to begin analysis while the run is still in progress. The higher cost per sample is acceptable because the total sample volume is low.

Tier 2: Regional or programmatic surveillance

This tier processes 50 to 500 samples per month. You need predictable throughput and a cost per sample that allows sustained operation. Illumina short-read sequencing fits this tier because you can multiplex many samples in a single run. The cost per sample decreases as you fill the flow cell, and the high base-level accuracy supports reliable resistance gene detection across a large sample set. The tradeoff is that you must batch samples to achieve the lowest cost, which introduces a delay between sample receipt and sequencing.

Tier 3: National or reference laboratory surveillance

This tier processes more than 500 samples per month. You need maximum throughput and the lowest possible cost per sample. Production-scale Illumina instruments are the standard choice because they support high multiplexing and generate large volumes of data per run. The bioinformatics infrastructure must be robust enough to process the data volume, and the laboratory needs dedicated personnel for data analysis and quality control.

Platform Selection Matrix

Use the following matrix to score each platform against your specific requirements. Assign a score of 1 to 5 for each criterion, where 5 means the platform fully meets your requirement and 1 means it does not.

The matrix is a starting point, not a final answer. You should adjust the weights based on your specific surveillance objectives. For example, if your program focuses on tracking the spread of resistance plasmids in a hospital setting, the resolution of plasmid structure may be more important than the cost per sample.

Decision Rules for Platform Selection

The following decision rules translate your scoring into a platform choice. These rules are based on the operational characteristics of the platforms and the requirements of AMR surveillance.

Rule 1: If your sample volume exceeds 100 samples per month and your primary objective is resistance gene detection, choose Illumina short-read sequencing.

This rule applies to most institutional and national surveillance programs. The cost per sample advantage of short-read sequencing becomes significant at this volume, and the high base-level accuracy ensures reliable detection of resistance genes. The limitation is that you will not be able to resolve the genomic context of the genes, but this is acceptable if your primary objective is to track resistance prevalence.

Rule 2: If your sample volume is below 50 samples per month and you need results within 24 to 48 hours, choose Oxford Nanopore sequencing.

This rule applies to outbreak investigations and point-of-care surveillance. The ability to sequence a single sample without batching is the critical advantage. The lower base-level accuracy is acceptable because you can confirm the results with phenotypic testing or with a follow-up short-read run.

Rule 3: If you need to resolve plasmid structures or complete bacterial genomes for a small number of priority samples, use long-read sequencing as a secondary platform.

This rule applies to reference laboratories that need to characterize the genomic context of resistance genes in detail. You can use short-read sequencing for the bulk of your samples and reserve long-read sequencing for the samples that require complete genomic resolution. This hybrid approach balances cost and genomic resolution.

Rule 4: If your program is in the early stages and you are uncertain about your long-term sample volume, start with a flexible platform that allows you to scale.

Oxford Nanopore offers flexibility because you can start with a small instrument and scale up as your sample volume increases. The consumable costs are based on the flow cells you use, so you can match your spending to your sample volume. This approach reduces the risk of investing in a high-throughput platform that you do not yet need.

Implementation Steps for the Framework

Step 1: Document your surveillance objectives and sample volume projections

Write down your expected sample volume for the next 12 months. Include the range of samples you expect to process per month, the types of organisms you will sequence, and the questions you need to answer. This documentation is the basis for your platform selection.

Step 2: Score each platform against your requirements

Use the platform selection matrix to score each platform against your specific requirements. Adjust the weights based on your surveillance objectives. For example, if turnaround time is critical for your program, increase the weight for the turnaround criterion.

Step 3: Calculate the total cost of ownership for your projected volume

The total cost of ownership includes the instrument, consumables, maintenance, and bioinformatics infrastructure. Calculate this cost for each platform at your projected sample volume. The cost per sample is not the only factor, but it is a useful metric for comparing platforms at scale.

Step 4: Validate the platform with a pilot run

Before you commit to a platform, run a pilot study with a small set of samples that have known resistance genes. The American Type Culture Collection provides authenticated biological materials that can be used for this purpose. The pilot run should confirm that the platform detects the resistance genes you expect to find and that the bioinformatics pipeline produces reliable results.

Step 5: Document the decision and the rationale

Record the scoring, the cost calculations, and the pilot results in a decision document. This document is important for future reference and for justifying the investment to your institution. The document should also include the criteria for revisiting the decision, such as a change in sample volume or the availability of new sequencing chemistry.

Records and Measurements for the Framework

The decision framework requires specific records to be useful. The following records should be maintained for each platform evaluation.

Sample volume records

Track the number of samples processed per month, the number of samples per sequencing run, and the number of runs per month. These records show whether your actual volume matches your projected volume and whether the platform is being used efficiently.

Cost records

Track the cost per sample for each run, including consumables, labor, and bioinformatics. The cost per sample should decrease as you increase the number of samples per run. If the cost per sample is not decreasing as expected, investigate the cause.

Turnaround records

Track the time from sample receipt to the final result for each run. The turnaround time includes the sequencing time, the bioinformatics analysis time, and the time for any confirmatory testing. The turnaround records show whether the platform meets your reporting deadlines.

Quality records

Track the quality metrics for each run, including the read length, the read accuracy, and the coverage. The quality records are essential for troubleshooting and for demonstrating the reliability of your results.

Common Failure Patterns in Platform Selection

Failure pattern 1: Choosing a platform based on instrument cost alone

Laboratories sometimes choose a platform because the instrument is inexpensive, without considering the consumable costs and the cost per sample. The total cost of ownership can be higher than expected if the consumable costs are high or if the throughput is lower than planned. The decision framework addresses this by requiring a total cost of ownership calculation.

Failure pattern 2: Choosing a platform based on the highest accuracy without considering the cost

High accuracy is valuable, but the highest accuracy is not always necessary for AMR surveillance. The accuracy of resistance gene detection depends on the bioinformatics pipeline as well as the platform. A platform with slightly lower accuracy may be sufficient if the pipeline is well validated. The decision framework addresses this by scoring the accuracy criterion against your specific requirements.

Failure pattern 3: Choosing a platform based on the longest read length without considering the throughput

Long reads are valuable for resolving genomic context, but they are not necessary for all surveillance programs. If your program is focused on detecting resistance genes in a large number of samples, the long-read platform may not be the most efficient choice. The decision framework addresses this by considering the throughput and the cost per sample.

Failure pattern 4: Choosing a platform without validating the bioinformatics pipeline

The sequencing platform is only one part of the workflow. The bioinformatics pipeline determines the accuracy of the resistance gene detection. If the pipeline is not validated, the results may be unreliable regardless of the platform. The decision framework addresses this by requiring a pilot run that includes the bioinformatics analysis.

When to Revisit the Platform Decision

The platform decision is not permanent. You should revisit the decision when the following conditions occur:

A change in sample volume

If your sample volume increases or decreases significantly, the cost per sample and the throughput requirements change. A platform that was cost-effective at a lower volume may not be cost-effective at a higher volume.

A change in surveillance objectives

If your surveillance objectives change, the platform may no longer be the best fit. For example, if you shift from detecting resistance genes to resolving the genomic context of resistance, you may need to add long-read sequencing.

A change in sequencing chemistry

The sequencing platforms are continuously improving. New chemistries may offer higher accuracy, longer reads, or lower costs. You should monitor the manufacturer announcements and evaluate whether the new chemistry changes the platform comparison.

A change in bioinformatics capacity

If your bioinformatics capacity changes, for example if you hire a bioinformatician or gain access to a high-performance computing cluster, the platform selection may change. The bioinformatics capacity affects the ability to process long-read data and to run complex analyses.

Professional Escalation Criteria

The decision framework is a practical tool, but it has limitations. If you encounter the following situations, escalate to a specialist or a public health authority.

Escalate to a bioinformatics specialist if you cannot validate the pipeline

If the pilot run does not produce reliable results, or if you cannot resolve the errors in the assembly, escalate to a bioinformatics specialist. The specialist can help you troubleshoot the pipeline and determine whether the platform is the source of the problem.

Escalate to a public health authority if you detect a resistance gene that is associated with a public health concern

The World Health Organization provides guidance on antimicrobial resistance surveillance and the public health response. If you detect a resistance gene that is associated with a public health concern, you may need to escalate to a public health authority. The escalation criteria depend on the organism and the resistance gene.

Escalate to a regulatory authority if you detect a resistance gene that is associated with a regulatory concern

The FDA provides guidance on laboratory methods for food microbiology, which may be relevant for foodborne pathogens. If you detect a resistance gene that is associated with a regulatory concern, you may need to escalate to a regulatory authority.

Safety and Quality Context

The decision framework should be implemented within the context of laboratory biosafety and quality management. The Centers for Disease Control and Prevention and the National Institutes of Health provide guidance on biosafety in microbiological and biomedical laboratories. The risk assessment should consider the organisms being sequenced and the procedures being performed.

The Clinical and Laboratory Standards Institute provides standards for antimicrobial susceptibility testing, which can be used to validate the phenotypic results. The sequencing results should be compared to the phenotypic results to confirm that the resistance genes you detect are associated with resistance. The American Society for Microbiology provides professional education and resources for laboratory professionals, which can help you develop the skills you need for the bioinformatics analysis.

Frequently Asked Questions

What is the difference between short-read and long-read sequencing?

Short-read sequencing produces reads of 150 to 300 base pairs, while long-read sequencing produces reads of 10,000 base pairs or more. Short reads are sufficient for detecting most resistance genes, while long reads provide the genomic context of the genes.

Which sequencing platform is best for AMR surveillance?

The best platform depends on your surveillance objectives. Illumina short-read sequencing is well suited for large-scale surveillance because of the low cost per sample. Oxford Nanopore long-read sequencing is well suited for understanding the genomic context of resistance.

How much does it cost to sequence a bacterial genome for AMR surveillance?

The cost per sample depends on the platform, the throughput, and the consumable costs. Illumina short-read sequencing typically has the lowest cost per sample for large-scale surveillance. Long-read sequencing is generally more expensive per sample.

How long does it take to sequence a bacterial genome?

The turnaround time depends on the platform and the sequencing chemistry. Illumina sequencing typically takes one to two days for a run. Oxford Nanopore sequencing can produce results in real-time, but the time depends on the throughput.

What bioinformatics skills are needed for AMR surveillance?

The bioinformatics skills needed include the ability to run the analysis pipeline, interpret the results, and troubleshoot problems. The American Society for Microbiology provides professional education resources for microbiology laboratory professionals.

How do I validate the sequencing platform for AMR surveillance?

You can validate the platform using reference strains with known resistance genes. The American Type Culture Collection provides authenticated biological materials for this purpose. The validation should confirm that the platform detects the resistance genes you expect to find.

What are the limitations of sequencing for AMR surveillance?

Sequencing identifies the genes that are present, but it does not confirm that the genes are expressed. The results should be validated with phenotypic testing to confirm the resistance. The Clinical and Laboratory Standards Institute provides standards for antimicrobial susceptibility testing.

How do I choose between Illumina and Oxford Nanopore?

The choice depends on your objectives. If you need to detect resistance genes in a large number of samples at a low cost per sample, Illumina is the appropriate choice. If you need to understand the genomic context of resistance, Oxford Nanopore may be necessary.

Related Bacteriology Guides

• Salmonellosis in Poultry: Serovar Surveillance, Antimicrobial Resistance, and Egg Safety
• Furunculosis in Salmonids: Detection and Prevention
• Avian Influenza in Australia: Epidemiology and Surveillance
• Avian Influenza in Wales: Surveillance and Control
• Avian Influenza: Global Surveillance and Pandemic Preparedness

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.
• Whole-genome sequencing for antimicrobial surveillance: species-specific quality thresholds and data evaluation from the network of the European Union Reference Laboratory for Antimicrobial Resistance genomic proficiency tests of 2021 and 2022.. mSystems, 2024.
• Contributions of Long-Read Sequencing for the Detection of Antimicrobial Resistance.. Pathogens (Basel, Switzerland), 2024.
• Comparison of Illumina and Oxford Nanopore Technology systems for the genomic characterization of Streptococcus pneumoniae.. Microbiology spectrum, 2025.
• Typhi Mykrobe: fast and accurate lineage identification and antimicrobial resistance genotyping directly from sequence reads for the typhoid fever agent Salmonella Typhi.. Genome medicine, 2025.

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