# Resistance Gene Databases Compared ## Quick Answer - Choose CARD for broad resistance mechanism coverage and ontology-based curation, ResFinder for targeted acquired resistance gene detection, and NCBI AMRFinderPlus for combined resistance and virulence determinant screening. - Match database choice to your specific analysis goal, sample type, and prediction tool compatibility, since each database has different strengths in coverage, curation, and output format. - No single database detects every resistance determinant, so results should be interpreted alongside phenotypic antimicrobial susceptibility testing for clinical or regulatory decisions. ## At a Glance | Database | Primary Focus | Curation Approach | Associated Tools | Best Use Case | |----------|--------------|-------------------|------------------|---------------| | CARD | Comprehensive resistance mechanisms including mutations and efflux pumps | Expert-curated with ontology-based structure | Resistance Gene Identifier (RGI) | Research on resistance mechanisms and novel determinants | | ResFinder | Acquired resistance genes in bacterial genomes | Curated from published literature and public databases | ResFinder web tool, PointFinder for mutations | Clinical and food safety surveillance for known acquired genes | | NCBI AMRFinderPlus | Resistance genes plus stress response and virulence determinants | Curated by NCBI staff with protein family models | AMRFinderPlus command line tool | Broad screening of bacterial genomes for multiple determinant types | ## Understanding Resistance Gene Databases Antimicrobial resistance represents a global public health challenge that requires coordinated surveillance and laboratory detection efforts. The World Health Organization identifies antimicrobial resistance as a threat to modern medicine, noting that resistant infections complicate treatment and increase healthcare costs. In response, whole genome sequencing has become a standard approach for characterizing bacterial isolates in research, clinical, and food safety laboratories. Resistance gene databases provide the reference information needed to interpret sequencing data and identify genetic determinants of resistance. The core function of a resistance gene database is to store curated sequences and associated metadata that allow researchers to compare their sequencing results against known resistance determinants. These databases differ in their scope, curation standards, and the types of resistance mechanisms they cover. Some focus exclusively on acquired resistance genes, while others include mutations in chromosomal genes that confer resistance. The choice of database affects the sensitivity and specificity of resistance prediction from whole genome sequencing data. Laboratory professionals working with bacterial isolates must understand the strengths and limitations of each database to select the appropriate tool for their specific application. The selection process involves considering the organism being studied, the resistance mechanisms of interest, the sequencing platform used, and the downstream analytical requirements. This article provides a detailed comparison of the three most widely used resistance gene databases, including their content, curation standards, associated tools, and performance characteristics. ## Database Content and Scope ### CARD The Comprehensive Antibiotic Resistance Database, known as CARD, focuses on providing a molecular reference for antimicrobial resistance. CARD organizes its content around the Antibiotic Resistance Ontology, which connects antibiotics, resistance mechanisms, and the genes that encode them. This ontology-based structure allows researchers to query the database using standardized terms and to understand the relationships between different resistance determinants. CARD includes acquired resistance genes, mutations in housekeeping genes that confer resistance, and resistance genes that are intrinsic to certain bacterial species. The database covers resistance mechanisms including enzymatic inactivation, target modification, efflux pumps, and reduced permeability. This broad scope makes CARD useful for researchers investigating the molecular basis of resistance, particularly when studying novel or poorly characterized resistance mechanisms. The database is curated by an interdisciplinary team that reviews published literature and public sequence data. Each entry includes information about the resistance mechanism, the antibiotic class affected, and the bacterial species in which the gene has been identified. CARD also provides information about the confidence level of each entry, which helps users assess the reliability of their results. ### ResFinder ResFinder is a database that focuses on acquired antimicrobial resistance genes. It was developed to support the detection of resistance genes in whole genome sequencing data, particularly for clinical and food safety applications. The database contains sequences of acquired resistance genes that have been experimentally verified to confer resistance to specific antibiotics. The scope of ResFinder is intentionally narrower than CARD, focusing on genes that are horizontally transferable and commonly found in clinically relevant bacteria. This focus allows ResFinder to maintain a curated set of high-confidence resistance genes that are directly relevant to phenotypic resistance. The database includes genes for resistance to major antibiotic classes, including beta-lactams, aminoglycosides, fluoroquinolones, and others. ResFinder is associated with a web-based tool that allows users to submit sequencing data and receive a report of detected resistance genes. The tool uses a BLAST-based approach to compare input sequences against the database and reports genes with a specified identity threshold. This workflow is designed to be accessible to researchers and laboratory professionals who may not have extensive bioinformatics experience. ### NCBI AMRFinderPlus The NCBI AMRFinderPlus database and tool were developed by the National Center for Biotechnology Information to support the detection of antimicrobial resistance genes in bacterial genomes. The database includes resistance genes as well as genes associated with stress response and virulence, providing a broader view of bacterial pathogenicity determinants. AMRFinderPlus uses protein family models to identify resistance genes, which allows it to detect genes that may not have high nucleotide identity to known sequences. The tool uses a hierarchical approach that first identifies protein families and then maps them to specific resistance genes. This approach can improve sensitivity for detecting novel variants of known resistance genes. The database includes acquired resistance genes, mutations in chromosomal genes that confer resistance, and stress response genes that may contribute to survival under antibiotic pressure. The inclusion of stress response genes is a distinguishing feature of AMRFinderPlus, as these genes may not directly confer resistance but can contribute to bacterial survival in the presence of antibiotics. ## Curation Standards and Quality Control ### CARD Curation Process CARD uses a structured curation process that involves manual review of published literature and public sequence data. Each entry is reviewed by curators who evaluate the evidence for the resistance gene and its mechanism. The database includes a confidence level for each entry, which reflects the strength of the evidence supporting the resistance function. The curation process for CARD includes regular updates to incorporate new resistance genes and mechanisms reported in the literature. The database is updated on a regular schedule, and the curation team monitors the scientific literature for new resistance determinants. This ongoing curation ensures that the database remains current with the evolving understanding of antimicrobial resistance. The ontology-based structure of CARD supports standardized queries and allows users to explore resistance mechanisms in a systematic way. The ontology connects genes to their mechanisms, the antibiotics they affect, and the organisms in which they are found. This structure is particularly useful for researchers who want to understand the broader context of a resistance gene. ### ResFinder Curation ResFinder curation focuses on maintaining a database of experimentally verified resistance genes. The database is curated by a team that reviews published studies and verifies that the resistance genes are associated with phenotypic resistance. This emphasis on experimental verification means that the database includes genes with a high confidence of conferring resistance. The curation process for ResFinder includes regular updates to add new resistance genes and to correct or remove entries that are no longer supported by evidence. The database is updated periodically, and the curation team monitors the literature for new resistance determinants. The focus on experimentally verified genes means that the database may not include genes that are predicted to confer resistance but have not been experimentally confirmed. The ResFinder database is designed to be compatible with the associated web tool, which uses a standardized approach for detecting resistance genes. The tool reports the identity and coverage of each detected gene, allowing users to assess the confidence of the detection. This transparency helps users interpret the results and identify potential false positives or false negatives. ### NCBI AMRFinderPlus Curation The NCBI AMRFinderPlus database is curated by NCBI staff who use a combination of manual curation and automated methods. The database includes protein family models that are used to identify resistance genes in bacterial genomes. The curation process involves reviewing the scientific literature and public sequence data to identify resistance determinants. The AMRFinderPlus database is updated regularly to include new resistance genes and to improve the protein family models. The NCBI curation team uses a structured process that includes validating the protein family models against known resistance genes and updating the database when new resistance determinants are identified. The database is designed to be comprehensive and to support the detection of resistance genes across a wide range of bacterial species. The AMRFinderPlus tool uses a hierarchical approach that first identifies protein families and then maps them to specific resistance genes. This approach can detect resistance genes that have diverged from known sequences, which is important for detecting novel resistance determinants. The tool also provides information about the confidence of each detection, helping users interpret the results. ## Associated Tools and Workflows ### CARD and RGI CARD is associated with the Resistance Gene Identifier, or RGI, which is a tool for predicting resistance from genome sequences. RGI uses the CARD database to identify resistance genes and mutations in genome sequences. The tool can be used with whole genome assemblies or raw sequencing reads, and it provides a report of detected resistance determinants. RGI uses a combination of protein homology and ontology-based approaches to identify resistance genes. The tool compares the input genome to the CARD database and reports genes that match known resistance determinants. RGI also identifies mutations in housekeeping genes that are known to confer resistance, which is a feature that is not available in all resistance detection tools. The RGI tool is available as a web-based application and as a command line tool for local use. The web-based version is accessible to users who do not have bioinformatics expertise, while the command line version allows for integration into larger analysis pipelines. The tool provides output in a structured format that can be used for further analysis. ### ResFinder and PointFinder ResFinder is associated with the ResFinder web tool, which is designed for the detection of acquired resistance genes. The tool accepts whole genome sequencing data and provides a report of detected resistance genes. The tool uses a BLAST-based approach to compare input sequences against the database and reports genes with high identity. The ResFinder web tool is designed to be user-friendly and accessible to laboratory professionals. Users can upload sequencing data in a variety of formats, and the tool provides results in a clear format. The tool also includes PointFinder, which is a companion tool for detecting mutations in chromosomal genes that confer resistance. The ResFinder tool provides information about the identity of each detected gene and the coverage of the gene sequence. This information helps users assess the confidence of the detection and to identify potential false positives. The tool also provides a summary of the resistance profile of the analyzed genome. ### NCBI AMRFinderPlus Tool The NCBI AMRFinderPlus tool is a command-line tool that can be used to detect resistance genes in bacterial genomes. The tool uses the AMRFinderPlus database and a protein-based approach to identify resistance determinants. The tool is designed for use in bioinformatics pipelines and can be integrated into larger analysis workflows. The AMRFinderPlus tool provides a detailed report of detected resistance genes, including the protein family, the specific resistance gene, and the confidence of the detection. The tool also identifies stress response genes and virulence determinants, which provides a broader view of the bacterial genome. The tool is designed to be fast and efficient, allowing it to be used with large datasets. The AMRFinderPlus tool is available for download from NCBI and can be used on a variety of operating systems. The tool requires a basic understanding of command-line interfaces, but it is designed to be accessible to users with some bioinformatics experience. The tool provides output in a structured format that can be used for further analysis. ## Performance in Predicting Phenotypic Resistance ### Sensitivity and Specificity The performance of resistance gene databases in predicting phenotypic resistance depends on the database content and the detection approach. Databases that include a comprehensive set of resistance genes are more likely to detect resistance determinants that are present in a genome. However, the presence of a resistance gene does not always correlate with phenotypic resistance, as gene expression and other factors can affect the phenotype. The sensitivity of a database refers to its ability to detect resistance genes that are present in a genome. A database with high sensitivity will detect most resistance genes, while a database with lower sensitivity may miss some genes. The specificity of a database refers to its ability to avoid false positives, which are genes that are detected but do not actually confer resistance. The performance of the databases varies depending on the organism and the resistance mechanism being studied. For example, a database that includes a comprehensive set of beta-lactamase genes will be more sensitive for detecting beta-lactam resistance than a database with a narrower set. The choice of database should be based on the specific resistance mechanisms that are relevant to the analysis. ### Correlation with Susceptibility Testing The correlation between resistance gene detection and phenotypic susceptibility testing is an important consideration for clinical and laboratory applications. Antimicrobial susceptibility testing, which is standardized by organizations such as the Clinical and Laboratory Standards Institute, provides a direct measurement of the susceptibility of an organism to antibiotics. Resistance gene detection can provide a genetic basis for the observed phenotype. The correlation between genotype and phenotype is not always perfect, as resistance can be caused by mechanisms that are not detected by the database. For example, resistance can be caused by mutations in genes that are not included in the database, or by efflux pumps that are not well characterized. The correlation can also be affected by the expression of resistance genes, which can vary depending on the genetic background of the organism. For clinical and regulatory decisions, the results of resistance gene detection should be interpreted in the context of antimicrobial susceptibility testing. The Clinical and Laboratory Standards Institute provides standards for antimicrobial susceptibility testing, and these standards should be used to confirm the phenotypic resistance of an isolate. The genetic detection can provide additional information about the resistance mechanism, but it should not be used as a substitute for phenotypic testing. ### Limitations of Prediction The prediction of phenotypic resistance from genomic data has several limitations that should be considered. The presence of a resistance gene does not always result in phenotypic resistance, as the gene may not be expressed or may not be functional. The absence of a resistance gene does not always mean that the organism is susceptible, as resistance can be conferred by mechanisms that are not included in the database. The databases are also limited by the current understanding of antimicrobial resistance. New resistance genes are constantly being discovered, and the databases may not include all of the resistance determinants that are present in a bacterial population. The databases are updated regularly, but the updates may not keep pace with the emergence of new resistance mechanisms. The prediction of phenotypic resistance from genomic data is a complex process that requires careful interpretation. The results of resistance gene detection should be interpreted in the context of the organism, the antibiotic, and the clinical or laboratory setting. The limitations of the prediction should be acknowledged, and the results should be confirmed with phenotypic testing when possible. ## Choosing the Right Database ### Consider the Analytical Question The choice of resistance database should be based on the specific analytical question being addressed. For researchers who are investigating the molecular basis of resistance, CARD provides a comprehensive view of resistance mechanisms and includes information about the ontology of resistance. For laboratory professionals who are identifying resistance genes in clinical or food safety isolates, ResFinder provides a focused set of experimentally verified genes. For researchers who are conducting a broad screening of bacterial genomes for resistance and other determinants, NCBI AMRFinderPlus provides a comprehensive approach that includes resistance genes, stress response genes, and virulence determinants. The choice of database should be based on the specific needs of the analysis, including the organism being studied, the resistance mechanisms of interest, and the downstream analytical requirements. The choice of database also depends on the type of data being analyzed. Some databases are designed for use with whole genome assemblies, while others can be used with raw sequencing reads. The choice of database should be compatible with the sequencing platform and the data analysis pipeline being used. ### Consider the Organism The choice of database should also consider the organism being analyzed. Some databases have a broader coverage of certain bacterial species, while others are more focused on specific groups. For example, a database that includes a comprehensive set of gram-negative resistance genes may be more useful for analyzing gram-negative isolates than a database with a narrower set. The databases also differ in their coverage of resistance mechanisms that are specific to certain organisms. For example, the resistance mechanisms of mycobacteria are different from those of gram-negative bacteria, and the databases may have different coverage of these mechanisms. The choice of database should be based on the organism being analyzed and the resistance mechanisms that are relevant. The databases are also updated to include new resistance genes that are discovered in different organisms. The choice of database should consider the current coverage of the database and the likelihood that the database includes the resistance genes that are relevant to the analysis. ### Consider the Tool Compatibility The choice of database should also consider the compatibility with the prediction tools that are being used. Each database is associated with a specific tool, and the choice of database should be compatible with the tool that is being used for the analysis. For example, CARD is associated with RGI, ResFinder is associated with the ResFinder web tool, and NCBI AMRFinderPlus is associated with the AMRFinder command line tool. The choice of tool should consider the user's bioinformatics expertise and the computational resources that are available. The web-based tools are designed to be accessible to users who do not have bioinformatics expertise, while the command line tools are designed for users who are comfortable with the command line interface. The choice of tool should also consider the size of the dataset and the computational resources that are required. ## Practical Implementation Steps ### Step 1: Define the Analytical Objective Before selecting a database, define the analytical objective and the specific questions that need to be answered. Consider the organism being analyzed, the resistance mechanisms of interest, and the downstream applications of the results. This will help to determine the database that is most appropriate for the analysis. ### Step 2: Evaluate the Database Content Evaluate the content of each database to determine whether it includes the resistance genes that are relevant to the analysis. Review the database documentation and the list of included genes to assess the coverage of the database. Consider the curation standards and the confidence level of the entries. ### Step 3: Select the Appropriate Tool Select the tool that is associated with the database and that is compatible with the analysis workflow. Consider the user's bioinformatics expertise and the computational resources that are available. The web-based tools are designed for users who do not have bioinformatics expertise, while the command line tools are designed for users who are comfortable with the command line. ### Step 4: Run the Analysis Run the analysis using the selected database and tool. Follow the tool documentation and the recommended parameters for the analysis. The tool will provide a report of detected resistance genes, including the identity and confidence of each detection. ### Step 5: Interpret the Results Interpret the results in the context of the analytical objective and the limitations of the database. The presence of a resistance gene does not always mean that the organism is resistant, and the absence of a resistance gene does not always mean that the organism is susceptible. The results should be interpreted in the context of the organism and the resistance mechanisms. ### Step 6: Confirm with Phenotypic Testing Confirm the results of the resistance gene detection with phenotypic testing when necessary. The Clinical and Laboratory Standards Institute provides standards for antimicrobial susceptibility testing, and these standards should be used to determine the phenotypic susceptibility of an isolate. The genetic detection can provide additional information about the resistance mechanism, but it should not be used as a substitute for phenotypic testing. ## Records and Measurements ### Documenting the Analysis Document the analysis process and the results of the resistance gene detection. This includes the database version, the tool version, the parameters used for the analysis, and the results of the detection. The documentation should be sufficient to allow the analysis to be reproduced and to allow the results to be interpreted in the context of the analytical objective. ### Recording the Database Version Record the database version and the date of the analysis. The databases are updated regularly, and the version of the database can affect the results of the analysis. The database version should be recorded to allow the results to be compared with the results of other analyses. ### Recording the Detection Results Record the detection results, including the resistance genes that were detected, the identity of each gene, and the confidence of the detection. The results should be recorded in a structured format that can be used for further analysis. The results should also be recorded in the context of the analytical objective and the limitations of the database. ## Common Failure Patterns ### False Positive Detections False positive detections occur when the database identifies a resistance gene that is not actually present in the genome or that does not confer resistance. This can occur when the database includes genes that are similar to other genes in the genome, or when the detection threshold is too low. False positives can be reduced by using a higher detection threshold and by confirming the results with phenotypic testing. ### False Negative Detections False negative detections occur when the database does not identify a resistance gene that is present in the genome. This can occur when the database does not include the resistance gene, or when the detection threshold is too high. False negatives can be reduced by using a database with a comprehensive coverage of resistance genes and by using a lower detection threshold. ### Database Version Differences The results of the resistance gene detection can vary depending on the version of the database that is used. The databases are updated regularly, and the new versions may include new resistance genes or may remove genes that are no longer supported by evidence. The results of the analysis should be interpreted in the context of the database version that was used. ## Safety and Regulatory Context ### Laboratory Biosafety The analysis of bacterial genomes for resistance genes should be conducted in accordance with the biosafety guidelines for the laboratory. The Biosafety in Microbiological and Biomedical Laboratories provides guidelines for the safe handling of biological materials, including bacterial isolates. The laboratory should follow the biosafety guidelines for the appropriate biosafety level for the organism being analyzed. ### Quality Control The quality of the resistance gene detection should be ensured by using authenticated biological materials and by following the quality control procedures for the laboratory. The American Type Culture Collection provides authenticated biological materials that can be used for quality control. The laboratory should also follow the quality control procedures for the sequencing and the analysis. ### Professional Standards The resistance gene detection should be conducted in accordance with the professional standards for the laboratory. The American Society for Microbiology provides professional education and laboratory practice resources for microbiologists. The laboratory should follow the professional standards for the analysis and the interpretation of the results. ## Common Failure Patterns ### Incomplete Database Coverage A common failure pattern is the use of a database that does not include the resistance genes that are relevant to the analysis. This can result in false negative detections and an incomplete view of the resistance profile of the organism. The database should be selected based on the coverage of the resistance genes that are relevant to the analysis. ### Inappropriate Detection Threshold The detection threshold is a key parameter that affects the sensitivity and specificity of the resistance gene detection. A threshold that is too high can result in false negatives, while a threshold that is too low can result in false positives. The detection threshold should be selected based on the analytical objective and the quality of the sequencing data. ### Lack of Phenotypic Confirmation The resistance gene detection should be confirmed with phenotypic testing when the results are used for clinical or laboratory decisions. The presence of a resistance gene does not always mean that the organism is resistant, and the absence of a resistance gene does not always mean that the organism is susceptible. The phenotypic testing should be used to confirm the results of the genetic detection. ## Safety and Escalation Criteria ### When to Escalate The results of the resistance gene detection should be escalated to a supervisor or a specialist when the results are unexpected or when the results have implications for clinical or laboratory decisions. The escalation should include the results of the analysis, the database and tool that were used, and the context of the analysis. ### When to Confirm with Phenotypic Testing The results of the resistance gene detection should be confirmed with phenotypic testing when the results are used for clinical or laboratory decisions. The phenotypic testing should be conducted in accordance with the standards of the Clinical and Laboratory Standards Institute. The results of the phenotypic testing should be used to confirm the results of the genetic detection. ### When to Seek Professional Guidance The results of the resistance gene detection should be discussed with a professional or a specialist when the results are not clear or when the results have implications for the treatment of an infection. The professional should be consulted to interpret the results in the context of the clinical or laboratory setting. ## A Practical Decision Framework for Database Selection Selecting a resistance gene database requires a structured approach that accounts for the specific analytical context, the organism under study, and the intended use of the results. A practical decision framework helps researchers move beyond general database descriptions and make choices that align with their laboratory workflow and reporting requirements. ### Step 1: Classify the Analysis Type Begin by classifying the analysis into one of three categories. The first category is discovery research, where the goal is to identify novel resistance mechanisms or to characterize the full resistance potential of an organism. The second category is routine surveillance, where the goal is to detect known resistance genes in clinical or food safety isolates for monitoring purposes. The third category is regulatory or clinical reporting, where the results will be used to support decisions about treatment or public health action. For discovery research, CARD provides the broadest coverage of resistance mechanisms, including mutations and efflux pumps that are not captured by other databases. For routine surveillance, ResFinder offers a focused set of experimentally verified acquired genes that are directly relevant to known resistance phenotypes. For regulatory or clinical reporting, NCBI AMRFinderPlus provides a comprehensive screening approach that includes resistance genes, stress response genes, and virulence determinants, which can be useful for characterizing the full pathogenicity profile of an isolate. ### Step 2: Assess the Organism and Sample Type The organism being analyzed should influence the database selection. Some databases have stronger coverage for specific bacterial groups. For example, a laboratory analyzing gram-negative isolates should verify that the selected database includes the relevant beta-lactamase genes and efflux pump systems common in those organisms. A laboratory analyzing gram-positive isolates should confirm that the database includes the appropriate cell wall synthesis target genes and modifying enzymes. The sample type also matters. For whole genome assemblies, all three databases can be used effectively. For raw sequencing reads, the compatibility of the associated tool with the read data should be verified. The RGI tool for CARD can accept raw reads, while the ResFinder web tool and AMRFinderPlus command line tool have specific input requirements that should be reviewed before starting the analysis. ### Step 3: Evaluate the Reporting Requirements The reporting requirements for the analysis should be defined before selecting a database. If the results will be used for clinical decisions, the database should be paired with phenotypic confirmation using antimicrobial susceptibility testing. The Clinical and Laboratory Standards Institute provides standards for antimicrobial susceptibility testing, and these standards should be followed to confirm the genetic findings. If the results will be used for research purposes, the database should be selected based on the depth of information needed about the resistance mechanism. CARD provides ontology-based information that connects genes to mechanisms and antibiotics, which is useful for understanding the biological context of the resistance. ResFinder provides a simpler output that focuses on the presence of known genes, which is useful for surveillance applications. ### Step 4: Assess the Computational Environment The computational environment available in the laboratory affects the choice of database and tool. The web-based tools for CARD and ResFinder are accessible to users who do not have command line experience. The AMRFinderPlus command line tool requires basic familiarity with command line interfaces but can be integrated into larger bioinformatics pipelines. The size of the dataset also affects the choice of tool. The web-based tools may have upload limits or processing time constraints for large datasets. The command line tool can be used with large datasets and can be run on a local server or high performance computing cluster. The laboratory should assess the available computational resources before selecting the tool. ### Step 5: Document the Decision The decision to use a specific database should be documented in the laboratory records. The documentation should include the analytical objective, the organism being analyzed, the database version, the tool version, and the parameters used for the analysis. This documentation allows the analysis to be reproduced and the results to be interpreted in the context of the database that was used. The documentation should also include the date of the analysis and the version of the database that was used. The databases are updated regularly, and the version of the database can affect the results of the analysis. The documentation should be sufficient to allow the results to be compared with the results of other analyses. ## A Record System for Database Selection A structured record system can help laboratories track the database selection process and the results of the analysis. The record system should include the following fields for each analysis: | Field | Description | |-------|-------------| | Analysis ID | Unique identifier for the analysis | | Date | Date of the analysis | | Analyst | Name of the person who performed the analysis | | Organism | Species of the organism being analyzed | | Sample Type | Whole genome assembly or raw reads | | Analytical Objective | Discovery, surveillance, or regulatory reporting | | Database Selected | CARD, ResFinder, or NCBI AMRFinderPlus | | Database Version | Version number of the database | | Tool Used | RGI, ResFinder web tool, or AMRFinderPlus | | Detection Threshold | The identity and coverage thresholds used | | Results Summary | Summary of the detected resistance genes | | Phenotypic Confirmation | Whether phenotypic testing was performed | This record system allows the laboratory to track the database selection process and to compare the results of different analyses. The records can be used to identify patterns in the database selection and to improve the decision making process over time. ### Troubleshooting Database Selection When the results of the analysis do not match the expected resistance profile, the laboratory should troubleshoot the database selection. The first step is to verify that the database version is current and that the database includes the resistance genes that are relevant to the organism being analyzed. The database documentation should be reviewed to confirm the coverage of the database. The second step is to verify that the detection threshold is appropriate for the analysis. A threshold that is too high can result in false negative detections, while a threshold that is too low can result in false positive detections. The threshold should be adjusted based on the quality of the sequencing data and the analytical objective. The third step is to verify that the tool is being used correctly. The tool documentation should be reviewed to confirm that the input data is in the correct format and that the parameters are set correctly. The tool should be run with a control sample to verify that the tool is working correctly. The fourth step is to confirm the results with phenotypic testing. The Clinical and Laboratory Standards Institute provides standards for antimicrobial susceptibility testing, and these standards should be used to confirm the results of the genetic detection. The phenotypic testing should be performed in accordance with the laboratory quality control procedures. ## Common Failure Patterns in Database Selection ### Selecting a Database Without Defining the Objective A common failure pattern is selecting a database without first defining the analytical objective. This can result in a database that does not provide the information that is needed for the analysis. The laboratory should define the analytical objective before selecting the database. ### Using an Outdated Database Version The databases are updated regularly to include new resistance genes and to improve the curation of existing entries. Using an outdated database version can result in false negative detections, as the database may not include the resistance genes that are present in the organism. The laboratory should record the database version and update the database regularly. ### Ignoring the Limitations of the Database The databases have limitations that should be acknowledged when interpreting the results. The presence of a resistance gene does not always correlate with phenotypic resistance, and the absence of a resistance gene does not always mean that the organism is susceptible. The laboratory should acknowledge the limitations of the database and confirm the results with phenotypic testing when necessary. ## Safety and Escalation Criteria ### When to Escalate the Database Selection The database selection should be escalated to a supervisor or a bioinformatics specialist when the analytical objective is not clear or when the results of the analysis are unexpected. The escalation should include the analytical objective, the organism being analyzed, and the database that was selected. The supervisor or specialist can provide guidance on the appropriate database for the analysis. ### When to Confirm with Phenotypic Testing The results of the resistance gene detection should be confirmed with phenotypic testing when the results are used for clinical or regulatory decisions. The phenotypic testing should be conducted in accordance with the standards of the Clinical and Laboratory Standards Institute. The results of the phenotypic testing should be used to confirm the results of the genetic detection. ### When to Seek Professional Guidance The results of the resistance gene detection should be discussed with a professional or a specialist when the results are not clear or when the results have implications for the treatment of an infection. The professional should be consulted to interpret the results in the context of the clinical or laboratory setting. The American Society for Microbiology provides professional education and laboratory practice resources for microbiologists, and these resources can be used to support the interpretation of the results. ## Frequently Asked Questions ### What is the difference between CARD and ResFinder? CARD provides a comprehensive view of resistance mechanisms, including acquired genes, mutations, and efflux pumps, while ResFinder focuses on acquired resistance genes that have been experimentally verified. CARD uses an ontology-based structure, while ResFinder uses a simpler database of verified genes. ### Which database is best for detecting resistance genes in whole genome sequencing? The best database depends on the analytical objective. CARD is useful for comprehensive resistance mechanism analysis, ResFinder is useful for targeted detection of known acquired genes, and NCBI AMRFinderPlus is useful for broad screening of resistance and other pathogenicity determinants. ### How do I choose between CARD and NCBI AMRFinderPlus? Choose CARD when you need a detailed understanding of resistance mechanisms and the ontology of resistance. Choose NCBI AMRFinderPlus when you need a broad screening of resistance genes, stress response genes, and virulence determinants in a single analysis. ### Can resistance gene databases predict phenotypic resistance? Resistance gene databases can predict the presence of resistance genes, but the presence of a gene does not always correlate with phenotypic resistance. The prediction should be confirmed with antimicrobial susceptibility testing, which is standardized by the Clinical and Laboratory Standards Institute. ### What are the limitations of resistance gene databases? The databases are limited by the coverage of known resistance genes and the correlation between the presence of a gene and phenotypic resistance. The databases are updated regularly, but they may not include all resistance mechanisms that are present in a bacterial population. ### How often are the resistance gene databases updated? The databases are updated regularly to include new resistance genes and to improve the curation of existing entries. The update frequency varies by database, and the database version should be recorded for the analysis. ### What is the role of antimicrobial susceptibility testing in resistance gene detection? Antimicrobial susceptibility testing provides a phenotypic measurement of the susceptibility of an isolate to antibiotics. The results of the resistance gene detection should be confirmed by antimicrobial susceptibility testing, which is standardized by the Clinical and Laboratory Standards Institute. ### How should the results of resistance gene detection be interpreted? The results should be interpreted in the context of the analytical objective, the organism being analyzed, and the limitations of the database. The presence of a resistance gene should be confirmed by phenotypic testing when the results are used for clinical or laboratory decisions. ## Related Bacteriology Guides - [Biological Foundation Models for Antimicrobial Peptide Discovery in Veterinary Pathogens](/knowledge/bacteria/general/biological-foundation-models-for-antimicrobial-peptide-discovery-in-veterinary-pathogens) - [Aeromonas hydrophila in Aquaculture: Virulence Factors and Antimicrobial Resistance](/knowledge/bacteria/aquatic-bacteria/aeromonas-hydrophila-aquaculture-virulence-factors-antimicrobial-resistance) - [Swine Bacterial Diseases: Colibacillosis, UTI, and Antimicrobial Resistance](/knowledge/bacteria/livestock-bacteria/swine-bacterial-diseases-colibacillosis-uti-antimicrobial-resistance) - [Avian Colibacillosis: Pathogenesis, Diagnosis, and Antimicrobial Resistance Patterns in Poultry](/knowledge/bacteria/avian-bacteria/avian-colibacillosis-pathogenesis-diagnosis-antimicrobial-resistance-poultry) - [Salmonellosis in Poultry: Serovar Surveillance, Antimicrobial Resistance, and Egg Safety](/knowledge/bacteria/avian-bacteria/salmonellosis-poultry-serovar-surveillance-antimicrobial-resistance-egg-safety) ## References and Further Reading - [Biosafety in Microbiological and Biomedical Laboratories](https://www.cdc.gov/labs/bmbl/index.html). CDC and NIH. - [Microbiology Resources](https://asm.org/). American Society for Microbiology. - [Antimicrobial Resistance](https://www.who.int/health-topics/antimicrobial-resistance). World Health Organization. - [Bacteriological Analytical Manual](https://www.fda.gov/food/laboratory-methods-food/bacteriological-analytical-manual-bam). U.S. Food and Drug Administration. - [Antimicrobial Susceptibility Testing](https://clsi.org/standards/products/microbiology). Clinical and Laboratory Standards Institute. - [Culture Collections](https://www.atcc.org/). American Type Culture Collection. - [NCBI Pathogen Detection](https://www.ncbi.nlm.nih.gov/pathogens). National Center for Biotechnology Information. - [Single-isolate genomic characterization and cgMLST allelic comparison of a bovine mastitis-associated Staphylococcus aureus ST151 isolate from Kazakhstan.](https://pubmed.ncbi.nlm.nih.gov/42597407). Frontiers in veterinary science, 2026. - [Integrating Phenotypic and Genomic Data with Machine Learning to Predict Antimicrobial Resistance and Identify Genetic Biomarkers in <,i>,E. coli<,/i>,.](https://doi.org/10.3390/ijerph23050561). 2026. - [Genome-Based Analysis of Chromosomal Colistin Non-Susceptibility in <,i>,Stenotrophomonas pavanii<,/i>, Isolated from the Phycosphere of <,i>,Pectinodesmus pectinatus<,/i>,.](https://doi.org/10.3390/antibiotics15050451). 2026. - [First comprehensive whole genome sequencing report of a multidrug-resistant <,i>,Serratia nematodiphila<,/i>, strain from Bangladesh.](https://doi.org/10.1016/j.nmni.2026.101743). 2026. > This article is educational and does not replace validated laboratory procedures, clinical diagnosis, institutional biosafety review, or public-health guidance.