AMRScan 2: provenance-recorded detection of antimicrobial resistance determinants in bacterial assemblies, with archived susceptibility data from NCBI Pathogen Detection
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Genomics Radio. Generated commentary on the latest computational biology and genomics papers.
Ines: We're looking at the paper: "AMRScan 2: provenance-recorded detection of antimicrobial resistance determinants in bacterial assemblies, with archived susceptibility data from NCBI Pathogen Detection".
Marcus: Here's the gist of the paper. AMRScan is a hybrid bioinformatics toolkit implemented in both R and Nextflow for the rapid and reproducible detection of antimicrobial resistance (AMR) genes from next-generation sequencing (NGS) data The toolkit enables users to identify AMR gene hits in sequencing reads by aligning them against reference databases such as CARD using BLAST (Altschul et al.
Ines: 1990) Toolkit Functionality AMRScan is designed to bridge the gap between lightweight exploratory analysis and production-ready surveillance pipelines It offers two usage modes tailored to user needs, specifically an R script for single-sample analysis and a Nextflow workflow for multi-sample batch processing The Nextflow implementation organizes the same logic into modular processes such as DownloadCARD, MakeBLASTdb, ConvertFASTQ, RunBLAST, and ParseResults Implementation Details The toolkit utilizes a hybrid approach where the R script encapsulates the entire pipeline in a linear format The Nextflow workflow organizes the logic into modular processes, ensuring reproducibility and scalability in HPC/cloud environments A shared R script is used for post-BLAST result summarization, which is documented and validated using mock NGS input Validation and Example Data The example data used to validate AMRScan was obtained from a study by Munim et al.
Marcus: (2024) on multidrug-resistant Klebsiella pneumoniae isolated from poultry in Noakhali, Bangladesh The detection process involved using the protein homolog model from the Comprehensive Antibiotic Resistance Database (CARD) version Broadstreet v4.0.1 A sample output summary shows hits such as OprA for pseudomonasaeruginosa and LAP-2 for Enterobacter cloacae Accessibility and Use Cases AMRScan addresses the need for a minimal, transparent, and reproducible toolkit suitable for small labs, clinical settings, or large-scale surveillance workflows The standalone R script remains accessible to users in data science, microbiology, and epidemiology This flexible dual-mode implementation ensures that AMRScan can serve both teaching/demo scenarios and production-grade bioinformatics pipelines Software Availability The source code for AMRScan is freely available on GitHub at https://github.com/biosciences/AMRScan The paper acknowledges collaborators at University of Sydney for feedback on early concepts The references include Altschul et al.
Ines: 1990, CARD 2017, and Munim et al.
Paper summary: Marcus: So wrapping up this discussion on AMRScan two: the paper emphasizes that this flexible dual-mode implementation lets you choose your path depending on whether you need a quick script or a scalable workflow <ref:2507.08062#pg1>.
Ines: The authors highlight that the toolkit is designed to serve both teaching and production-grade pipelines, which really shows they’re thinking about different user needs from the start.
Yuki: From my perspective, it suggests that by having this provenance recording and linking to archived susceptibility data, we can build better tools for understanding how resistance determinants evolve within bacterial populations over time.
Marcus: It’s about making sure that when you identify a resistance gene, you don't just have the sequence; you also have the historical context of its known susceptibility.
Ines: The implication is that this approach makes it easier for labs, whether they are small or large surveillance teams, to move from just finding a hit to actually understanding the clinical significance of that hit.
Conclusion: Ines: So we’ve seen how AMRScan two works, and now we’re looking at what this paper actually says about it.
Marcus: The title is pretty descriptive, isn't it? "AMRScan two: provenance-recorded detection of antimicrobial resistance determinants in bacterial assemblies."
Ines: Right. It’s not just detecting a gene anymore; it’s recording where that gene came from and linking it to old susceptibility data.
Marcus: That linkage is key, I think. They're taking these new sequencing reads and tracing the resistance genes back to known samples from NCBI Pathogen Detection.
Yuki: From a population genetics view, that provenance information allows us to see if this specific resistance mechanism is spreading in a real outbreak or if it’s just random noise.
Ines: Exactly. We move beyond just saying "this strain has this gene" to saying "this gene came from sample number four from last year."
Marcus: And for the cohort side, that means we can finally see if batch effects are skewing our resistance estimates because we have this external ground truth data to compare against.
Yuki: It supports the idea that these resistance genes aren't just isolated events; they are part of a history within the bacterial population.
Ines: The implication is that surveillance moves from being purely reactive to being proactive, allowing us to predict which strains might become high-risk based on their genetic lineage.
Marcus: And if we can do that, it changes how we design our screening pipelines for large scale monitoring.
Yuki: It suggests that understanding the history of resistance determinants is just as important as identifying the genes themselves.
University of Sydney
q-bio.GN
Submitted: 2025-07-10
Updated: 2026-10-31
Comments: 10 pages
Code: https://github.com/biosciences/AMRScan
License: http://creativecommons.org/licenses/by-nc-sa/4.0/
Importance score: 91/100
The gist: AMRScan is a hybrid bioinformatics toolkit implemented in both R and Nextflow for the rapid and reproducible detection of antimicrobial resistance (AMR) genes from next-generation sequencing (NGS)…
Key concepts
- Provenance-recorded detection
- This means tracking the exact source or history of an AMR gene found in a bacterial sample. Instead of just knowing a gene is present, this method records where it originated within the bacterial genome assembly, providing traceability for resistance mechanisms.
- Antimicrobial Resistance (AMR) determinants
- These are specific genes within bacteria that allow them to survive exposure to antibiotics. AMRScan 2 focuses on identifying these genes, which are crucial because they determine whether a bacterial strain is resistant or susceptible to various drugs.
- Archived susceptibility data from NCBI Pathogen Detection
- This refers to historical records stored in the NCBI database detailing how different bacterial strains have responded to specific antibiotics. Integrating this data helps link the detected AMR genes directly to known resistance profiles, providing context for their clinical relevance.
Terminology
Summary
AMRScan is a hybrid bioinformatics toolkit implemented in both R and Nextflow for the rapid and reproducible detection of antimicrobial resistance (AMR) genes from next-generation sequencing (NGS) data The toolkit enables users to identify AMR gene hits in sequencing reads by aligning them against reference databases such as CARD using BLAST (Altschul et al. p. 0 AMRScan: A hybrid R and Nextflow toolkit for rapid antimicrobial
resistance gene detection from sequencing data
Kaitao Lai1
2025-07-08
1 University of Sydney
1 Summary
AMRScan is a hybrid bioinformatics toolkit implemented in both R and Nextflow for the rapid and
reproducible detection of antimicrobial resistance (AMR) genes from next-generation sequencing (NGS) data. The toolkit enables users to identify AMR gene hits in sequencing reads by aligning them against reference
databases such as CARD using BLAST (Altschul et al.Where the paper says it
Toolkit Functionality
AMRScan is designed to bridge the gap between lightweight exploratory analysis and production-ready surveillance pipelines It offers two usage modes tailored to user needs, specifically an R script for single-sample analysis and a Nextflow workflow for multi-sample batch processing The Nextflow implementation organizes the same logic into modular processes such as DownloadCARD, MakeBLASTdb, ConvertFASTQ, RunBLAST, and ParseResults
Implementation Details
The toolkit utilizes a hybrid approach where the R script encapsulates the entire pipeline in a linear format The Nextflow workflow organizes the logic into modular processes, ensuring reproducibility and scalability in HPC/cloud environments A shared R script is used for post-BLAST result summarization, which is documented and validated using mock NGS input
Validation and Example Data
The example data used to validate AMRScan was obtained from a study by Munim et al. (2024) on multidrug-resistant Klebsiella pneumoniae isolated from poultry in Noakhali, Bangladesh The detection process involved using the protein homolog model from the Comprehensive Antibiotic Resistance Database (CARD) version Broadstreet v4. p. 1 For antimicrobial resistance gene detection, we used the protein homolog model from the Comprehensive
Antibiotic Resistance Database (CARD) (Jia et al. 2017), version Broadstreet v4.Where the paper says it
Accessibility and Use Cases
AMRScan addresses the need for a minimal, transparent, and reproducible toolkit suitable for small labs, clinical settings, or large-scale surveillance workflows The standalone R script remains accessible to users in data science, microbiology, and epidemiology This flexible dual-mode implementation ensures that AMRScan can serve both teaching/demo scenarios and production-grade bioinformatics pipelines
Software Availability
The source code for AMRScan is freely available on GitHub at https://github.com/biosciences/AMRScan The paper acknowledges collaborators at University of Sydney for feedback on early concepts The references include Altschul et al. p. 1 com/biosciences/AMRScan
7 Acknowledgements
The author thanks collaborators at University of Sydney for feedback on early concepts. References
Altschul, Stephen F, Warren Gish, Webb Miller, Eugene W Myers, and David J Lipman. For antimicrobial resistance gene detection, we used the protein homolog model from the Comprehensive
Antibiotic Resistance Database (CARD) (Jia et al. 2017), version Broadstreet v4.Where the paper says it
Where the paper says it
--- Page 3 ---
Jia, Baofeng, Aliyu R Raphenya, Brian Alcock, Nicholas Waglechner, Peng Guo, Katelyn K Tsang, Brendan A Lago, et al. 2017. “CARD 2017: Expansion and Model-Centric Curation of the Comprehensive Antibiotic Resistance Database. p. 2 Jia, Baofeng, Aliyu R Raphenya, Brian Alcock, Nicholas Waglechner, Peng Guo, Katelyn K Tsang, Brendan A
Lago, et al. “CARD 2017: Expansion and Model-Centric Curation of the Comprehensive Antibiotic
Resistance Database. “CARD 2017: Expansion and Model-Centric Curation of the Comprehensive Antibiotic
Resistance Database. ” Nucleic Acids Research 45 (D1): D566–73. ” Journal of Molecular Biology 215 (3): 403–10. https://doi. ” Journal of Molecular Biology 215 (3): 403–10. https://doi. ” Journal of Molecular Biology 215 (3): 403–10. https://doi.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Munim, Md Adnan, Afroza Akter Tanni, Md Mobarok Hossain, Kallyan Chakma, Adnan Mannan, SM Rafiqul Islam, Jully Gogoi Tiwari, and Shipan Das Gupta. 2024. “Whole Genome Sequencing of MultidrugResistant Klebsiella Pneumoniae from Poultry in Noakhali, Bangladesh: Assessing Risk of Transmission to Humans in a Pilot Study. p. 2 Munim, Md Adnan, Afroza Akter Tanni, Md Mobarok Hossain, Kallyan Chakma, Adnan Mannan, SM Rafiqul
Islam, Jully Gogoi Tiwari, and Shipan Das Gupta. “Whole Genome Sequencing of MultidrugResistant Klebsiella Pneumoniae from Poultry in Noakhali, Bangladesh: Assessing Risk of Transmission
to Humans in a Pilot Study. “Whole Genome Sequencing of MultidrugResistant Klebsiella Pneumoniae from Poultry in Noakhali, Bangladesh: Assessing Risk of Transmission
to Humans in a Pilot Study. ” Comparative Immunology, Microbiology and Infectious Diseases 114: 102246.Where the paper says it
Where the paper says it
Zankari, Ea, Henrik Hasman, Salvatore Cosentino, Martin Vestergaard, Simon Rasmussen, Ole Lund, Frank M Aarestrup, and Mette V Larsen. 2012. “Identification of Acquired Antimicrobial Resistance Genes. p. 2 Zankari, Ea, Henrik Hasman, Salvatore Cosentino, Martin Vestergaard, Simon Rasmussen, Ole Lund, Frank
M Aarestrup, and Mette V Larsen. “Identification of Acquired Antimicrobial Resistance Genes. “Identification of Acquired Antimicrobial Resistance Genes. ”
Journal of Antimicrobial Chemotherapy 67 (11): 2640–44.Where the paper says it
Where the paper says it
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Improved AI systems can perform rapid antimicrobial resistance gene detection by aligning sequencing reads against reference databases such as CARD using BLAST, as described in Section 1 of AMRScan.
-
The R implementation allows for a
concise, script-based approach suitable for single-sample analysis
enabling quick local tests and educational use of AMR gene identification. -
The Nextflow implementation enables
reproducible, scalable workflows for multi-sample batch processing in high-performance computing (HPC) and containerized environments.
-
The Nextflow workflow organizes logic into modular processes: DownloadCARD, MakeBLASTdb, ConvertFASTQ, RunBLAST, and ParseResults to ensure automated pipeline execution.
-
The toolkit bridges the gap between
lightweight exploratory analysis and production-ready surveillance pipelines,
making it suitable for both research and public health genomics applications.
Abstract
Studies that relate bacterial genomes to antimicrobial susceptibility test (AST) results need determinant calls and phenotypes that can be traced to the exact genome, software, database and source export they came from. AMRScan 2 is a rewrite of the AMRScan toolkit for this purpose. Its Nextflow workflow runs NCBI AMRFinderPlus on genome assemblies in a pinned container, writes a provenance record for every sample with the command, the software version and the SHA-256 hashes of the input, of every database file and of the report, and converts each report into a table with a fixed schema that keeps every original column. The conversion step stops if a report no longer matches its provenance record. A second tool archives AST and isolate tables exported from NCBI Pathogen Detection, checks each export against an independent count, and builds isolate and endpoint tables offline that can be rebuilt and compared byte for byte. AMRScan does not infer phenotype from determinants; calls and phenotypes are kept as separate layers and joined by assembly accession. The caller wrapper was used to call 10,521 public Escherichia coli assemblies on a workstation without a failed call, and a run of 1,000 of these assemblies through the workflow on an HPC system gave byte-identical reports. The workflow follows nf-core conventions and is tested with unit tests, nf-test and continuous integration. This version replaces the 2025 description of AMRScan 1, a BLAST search against CARD, which is kept unchanged in the repository.
Related papers
- SeekRBP: Leveraging Sequence-Structure Integration with Reinforcement Learning for Receptor-Binding Protein Identification
- Genome-Factory: A Library for Tuning, Deploying, and Interpreting Genomic Foundation Models
- AVA: Agentic Variant Adjudication for Automated, Auditable Read-Level Review in Cancer Diagnostic Sequencing
- C3P: Contrastive promoter-protein pretraining yields representations capturing bacterial gene regulation
- Leveraging genomic deep learning models for the prediction of non-coding variant effects
- Adaptive Allele-Haplotype Incidence Indexing for Pangenome Queries: Representation, Reconstruction, and Benchmarking