Abstract
Whole-genome sequencing (WGS) data are an invaluable resource for understanding antimicrobial resistance (AMR) mechanisms. However, WGS data are high-dimensional and the lack of standardized genomic representations is a key barrier to AMR phenotype prediction. To fully explore these high-resolution data, we propose AMR-GNN, a graph deep learning-based framework that integrates multiple genomic representations with graph neural networks (GNN) to enable AMR phenotype prediction from genomic sequence data. We test AMR-GNN with Pseudomonas aeruginosa, a clinically relevant Gram-negative bacterial pathogen known for its complex AMR mechanisms. We present AMR-GNN as a proof-of-concept framework designed to address several key problems in AMR phenotype prediction with data-driven machine learning (ML) approaches, including using multiple genomic representations to enhance performance, to mitigate the influence of clonal relationships and to identify informative biomarkers to provide explainability. Follow-up validation on the largest publicly available dataset spanning both Gram-negative and Gram-positive pathogens highlights AMR-GNN’s broad applicability in detecting AMR in diverse and clinically relevant pathogen-drug combinations.
Similar content being viewed by others
Data availability
The in-house sequence data generated in this study have been deposited in the NCBI BioProject ID PRJNA1220180. The Sequence Read Archive (SRA) accession numbers and AST data of all isolates can be found in Supplementary Data 1.
Code availability
The source code of this study is available at https://github.com/andyvng/amr-gnn. A citable version with a DOI is available on Zenodo (https://doi.org/10.5281/zenodo.18500983)69.
References
Naghavi, M. et al. Global burden of bacterial antimicrobial resistance 1990-2021: a systematic analysis with forecasts to 2050. Lancet 404, 1199–1226 (2024).
Macesic, N., Uhlemann, A. C. & Peleg, A. Y. Multidrug-resistant Gram-negative bacterial infections. Lancet 405, 257–272 (2025).
Kim, J. eeI. et al. Machine learning for antimicrobial resistance prediction: current practice, limitations, and clinical perspective. Clin. Microbiol. Rev. 35, e00179–21 (2022).
Macesic, N., Polubriaginof, F. & Tatonetti, N. P. Machine learning: novel bioinformatics approaches for combating antimicrobial resistance. Curr. Opin. Infect. Dis. 30, 511–517 (2017).
Hu, K. et al. Assessing computational predictions of antimicrobial resistance phenotypes from microbial genomes. Brief. Bioinform. 25, bbae206 (2024).
Green, A. G. et al. A convolutional neural network highlights mutations relevant to antimicrobial resistance in Mycobacterium tuberculosis. Nat. Commun. 13, 3817 (2022).
Macesic, N. et al. Predicting phenotypic polymyxin resistance in Klebsiella pneumoniae through machine learning analysis of genomic data. mSystems 5, e00656–19 (2020).
Moradigaravand, D. et al. Prediction of antibiotic resistance in Escherichia coli from large-scale pan-genome data. PLoS Comput. Biol. 14, e1006258 (2018).
Ren, Y. et al. Prediction of antimicrobial resistance based on whole-genome sequencing and machine learning. Bioinformatics 38, 325–334 (2021).
Acosta, J. N., Falcone, G. J., Rajpurkar, P. & Topol, E. J. Multimodal biomedical AI. Nat. Med. 28, 1773–1784 (2022).
Khaledi, A. et al. Predicting antimicrobial resistance in Pseudomonas aeruginosa with machine learning-enabled molecular diagnostics. EMBO Mol. Med. 12, e10264 (2020).
Johnson, R., Li, M. M., Noori, A., Queen, O. & Zitnik, M. Graph artificial intelligence in medicine. Ann. Rev. Biomed. Data Sci. 7, 345–368 (2024).
Břinda, K. et al. Rapid inference of antibiotic resistance and susceptibility by genomic neighbour typing. Nat. Microbiol. 5, 455–464 (2020).
Li, M. M., Huang, K. & Zitnik, M. Graph representation learning in biomedicine and healthcare. Nat. Biomed. Eng. 6, 1353–1369 (2022).
Zhang Z et al. Leveraging graph neural networks for mic prediction in antimicrobial resistance studies. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. 2024, 1–4 (2024).
Langendonk, R. F., Neill, D. R. & Fothergill, J. L. The building blocks of antimicrobial resistance in Pseudomonas aeruginosa: implications for current resistance-breaking therapies. Front. Cell. Infect. Microbiol. 11, 665759 (2021).
Miller, W. R. & Arias, C. A. ESKAPE pathogens: antimicrobial resistance, epidemiology, clinical impact and therapeutics. Nat. Rev. Microbiol. 22, 598–616 (2024).
Kos, V. N. et al. The resistome of Pseudomonas aeruginosa in relationship to phenotypic susceptibility. Antimicrob. Agents Chemother. 59, 427–436 (2014).
Barrio-Tofiño, ED. et al. Genomics and susceptibility profiles of extensively drug-resistant Pseudomonas aeruginosa isolates from Spain. Antimicrob. Agents Chemother. 61, e01589–17 (2017).
Barrio-Tofiño, E. D. et al. Spanish nationwide survey on Pseudomonas aeruginosa antimicrobial resistance mechanisms and epidemiology. J. Antimicrob. Chemother. 74, 1825–1835 (2019).
Cortes-Lara, S. et al. Predicting Pseudomonas aeruginosa susceptibility phenotypes from whole genome sequence resistome analysis. Clin. Microbiol. Infect. 27, 1631–1637 (2021).
Teo, JQ-M. et al. Ceftolozane/Tazobactam resistance and mechanisms in carbapenem-nonsusceptible Pseudomonas aeruginosa. mSphere 6, e01026–20 (2021).
Rebelo, A. R. et al. One day in Denmark: comparison of phenotypic and genotypic antimicrobial susceptibility testing in bacterial isolates from clinical settings. Front. Microbiol. ume 13, 804627 (2022).
Stanton, R. A. et al. Whole-genome sequencing reveals diversity of carbapenem-resistant Pseudomonas aeruginosa collected through CDC’s emerging infections program, United States, 2016-2018. Antimicrob. Agents Chemother. 66, e00496–22 (2022).
Sastre-Femenia, M. À et al. Pseudomonas aeruginosa antibiotic susceptibility profiles, genomic epidemiology and resistance mechanisms: a nation-wide five-year time lapse analysis. Lancet Reg. Health Eur. 34, 100736 (2023).
European Committee on Antimicrobial Susceptibility Testing. Breakpoint tables for interpretation of MICs and zone diameters, version 15.0, 2025. Retrieved September 10, 2025, from https://www.eucast.org/clinical_breakpoints (2025).
Wang, T. et al. MOGONET integrates multi-omics data using graph convolutional networks allowing patient classification and biomarker identification. Nat. Commun. 12, 3445 (2021).
Liu, Z. et al. Efficient low-rank multimodal fusion with modality-specific factors. Preprint at https://doi.org/10.48550/arXiv.1806.00064 (2018).
Deschavanne, P. J., Giron, A., Vilain, J., Fagot, G. & Fertil, B. Genomic signature: characterization and classification of species assessed by chaos game representation of sequences. Mol. Biol. Evol. 16, 1391–1399 (1999).
Almeida, J. S., Carriço, J. A., Maretzek, A., Noble, P. A. & Fletcher, M. Analysis of genomic sequences by chaos game representation. Bioinformatics 17, 429–437 (2001).
Earle, S. G. et al. Identifying lineage effects when controlling for population structure improves power in bacterial association studies. Nat. Microbiol. 1, 16041 (2016).
Liu, Y. et al. NeutronSketch: an in-depth exploration of redundancy in large-scale graph neural network training. Knowl. Based Syst. 309, 112786 (2025).
Rong, Y., Huang, W., Xu, T., Huang, J. DropEdge: towards deep graph convolutional networks on node classification. international conference on learning representations (ICLR). Preprint at https://doi.org/10.48550/arXiv.1907.10903 (2020).
Ransom, E. M., Potter, R. F., Dantas, G. & Burnham, C.-A. D. Genomic prediction of antimicrobial resistance: ready or not, here it comes!. Clin. Chem. 66, 1278–1289 (2020).
Madden, D. E. et al. Keeping up with the pathogens: improved antimicrobial resistance detection and prediction from Pseudomonas aeruginosa genomes. Genome Med. 16, 78 (2024).
Kim, J. et al. VAMPr: VAriant mapping and prediction of antibiotic resistance via explainable features and machine learning. PLoS Comput. Biol. 16, e1007511 (2020).
Olson, R. D. et al. Introducing the bacterial and viral bioinformatics resource center (BV-BRC): a resource combining PATRIC, IRD and ViPR. Nucleic Acids Res. 51, D678–D689 (2023).
Wattam, A. R. et al. Improvements to PATRIC, the all-bacterial bioinformatics database and analysis resource center. Nucleic Acids Res. 45, D535–D542 (2017).
Sundararajan, M., Taly, A., Yan, Q. Axiomatic attribution for deep networks. Preprint at https://doi.org/10.48550/arXiv.1703.01365 (2017).
Drlica, K. & Zhao, X. DNA gyrase, topoisomerase IV, and the 4-quinolones. Microbiol. Mol. Biol. Rev. 61, 377–392 (1997).
Tamber, S., Ochs, M. M. & Hancock, R. E. W. Role of the novel OprD family of porins in nutrient uptake in Pseudomonas aeruginosa. J. Bacteriol. 188, 45–54 (2006).
Chalhoub, H. et al. Mechanisms of intrinsic resistance and acquired susceptibility of Pseudomonas aeruginosa isolated from cystic fibrosis patients to temocillin, a revived antibiotic. Sci. Rep. 7, 40208 (2017).
Bolard, A., Plésiat, P., Jeannot, K. Mutations in gene fusA1 as a novel mechanism of aminoglycoside resistance in clinical strains of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 62, e01835–17 (2018).
Hyun, J. C., Kavvas, E. S., Monk, J. M. & Palsson, B. O. Machine learning with random subspace ensembles identifies antimicrobial resistance determinants from pan-genomes of three pathogens. PLoS Comput. Biol. 16, e1007608 (2020).
Liu, B. et al. Direct prediction of carbapenem resistance in Pseudomonas aeruginosa by whole genome sequencing and metagenomic sequencing. J. Clin. Microbiol. 61, e00617-23 (2023).
Leekitcharoenphon, P., Nielsen, E. M., Kaas, R. S., Lund, O. & Aarestrup, F. M. Evaluation of whole genome sequencing for outbreak detection of Salmonella enterica. PLoS ONE 9, e87991 (2014).
Dötsch, A. et al. The Pseudomonas aeruginosa transcriptional landscape is shaped by environmental heterogeneity and genetic variation. mBio 6, e00749 (2015).
Kocsis, B., Gulyás, D. & Szabó, D. Diversity and distribution of resistance markers in Pseudomonas aeruginosa international high-risk clones. Microorganisms 9, 359 (2021).
Nguyen, M., Olson, R., Shukla, M., VanOeffelen, M. & Davis, J. J. Predicting antimicrobial resistance using conserved genes. PLoS Comput. Biol. 16, e1008319 (2020).
Nguyen, H.-A. et al. Complex pathways to ceftolozane-tazobactam resistance in clinical Pseudomonas aeruginosa isolates: a genomic epidemiology study. Clin. Microbiol. Infect. 32, 110–117 (2026).
Weis, C. et al. Direct antimicrobial resistance prediction from clinical MALDI-TOF mass spectra using machine learning. Nat. Med. 28, 164–174 (2022).
Nguyen, H.-A. et al. Predicting Pseudomonas aeruginosa drug resistance using artificial intelligence and clinical MALDI-TOF mass spectra. mSystems 9, e00789-24 (2024).
Wick, R. R., Judd, L. M., Gorrie, C. L. & Holt, K. E. Unicycler: resolving bacterial genome assemblies from short and long sequencing reads. PLoS Comput. Biol. 13, e1005595 (2017).
Holley, G. & Melsted, P. Bifrost: highly parallel construction and indexing of colored and compacted de Bruijn graphs. Genome Biol. 21, 249 (2020).
Danecek, P. et al. Twelve years of SAMtools and BCFtools. GigaScience 10, giab008 (2021).
Lees, JA. et al. Improved prediction of bacterial genotype-phenotype associations using interpretable pangenome-spanning regressions. mBio 11, e01344–20 (2020).
Jin, C. et al. Predicting antimicrobial resistance in E. coli with discriminative position fused deep learning classifier. Comput. Struct. Biotechnol. J. 23, 559–565 (2024).
Drouin, A. et al. Predictive computational phenotyping and biomarker discovery using reference-free genome comparisons. BMC Genom. 17, 754 (2016).
Lees, J. A., Galardini, M., Bentley, S. D., Weiser, J. N. & Corander, J. pyseer: a comprehensive tool for microbial pangenome-wide association studies. Bioinformatics 34, 4310–4312 (2018).
Bengio, Y. Practical Recommendations for Gradient-Based Training of Deep Architectures. in Neural Networks: Tricks of the Trade: 2nd Edn (eds Montavon, G., Orr, GB., Müller, K-R.) Lecture Notes in Computer Science. 437–478 (Springer, 2012).
Masters, D., Luschi, C. Revisiting small batch training for deep neural networks. Preprint at https://doi.org/10.48550/arXiv.1804.07612 (2018).
Clinical and Laboratory Standards Institute. CLSI M52: Verification of Commercial Microbial Identification and Antimicrobial Susceptibility Testing Systems. Wayne, PA, Clinical and Laboratory Standards Institute (2015).
Kipf, TN., Welling, M. Semi-supervised classification with graph convolutional networks. Preprint at https://doi.org/10.48550/arXiv.1609.02907 (2016).
Schwengers, O. et al. Bakta: rapid and standardized annotation of bacterial genomes via alignment-free sequence identification. Microbial. Genom. 7, https://doi.org/10.1099/mgen.0.000685 (2021).
Alcock, B. P. et al. CARD 2023: expanded curation, support for machine learning, and resistome prediction at the comprehensive antibiotic resistance database. Nucleic Acids Res. 51, D690–D699 (2023).
Kokhlikyan N, et al. Captum: a unified and generic model interpretability library for PyTorch. Preprint at https://doi.org/10.48550/arXiv.2009.07896 (2020).
Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. Basic local alignment search tool. J. Mol. Biol. 215, 403–410 (1990).
Lundberg, S., Lee, S-I. A unified approach to interpreting model predictions. Preprint at https://doi.org/10.48550/arXiv.1705.07874 (2017).
Nguyen, H-A. et al. AMR-GNN: A multi-representation graph neural network framework to enable genomic antimicrobial resistance prediction. Code for AMR-GNN. Preprint at https://doi.org/10.1101/2025.07.24.666581 (2026).
Varadarajan, A. R. et al. An integrated model system to gain mechanistic insights into biofilm-associated antimicrobial resistance in Pseudomonas aeruginosa MPAO1. npj Biofilms Microbiomes 6, 46 (2020).
Patel, Y., Soni, V., Rhee, K. Y. & Helmann, J. D. Mutations in rpoB that confer rifampicin resistance can alter levels of peptidoglycan precursors and affect β-lactam susceptibility. mBio 14, e03168–22 (2023).
Acknowledgements
This work was supported by the National Health and Medical Research Council of Australia (Emerging Leader 1 Fellowship APP1176324 to N.M. and Practitioner Fellowship APP1117940 to A.Y.P) and the Australian Medical Research Future Fund (FSPGN000048). The funders had no role in study design, data collection, and interpretation, or the decision to submit the work for publication. This work was supported by Monash eResearch Center, including the M3 service. This work was also supported by use of the Nectar Research Cloud, a collaborative Australian research platform supported by the National Collaborative Research Infrastructure Strategy (NCRIS)-funded Australian Research Data Commons (ARDC).
Author information
Authors and Affiliations
Contributions
HA.N, A.Y.P., and N.M. conceived the study. J.A.W. designed and supervised sampling and collection of bacterial isolates. J.A.W., L.V.B., G.Z.B, R.T., N.Q.D., and M.H.P. collected the bacterial isolates, performed bacterial characterization, conducted whole genome sequencing, and broth microdilution. HA.N. preprocessed data. HA.N., J.S., D.L.D., A.G.G., Z.W., X.W., and N.M. conceptualized machine learning analyses. HA.N. developed machine learning models and evaluated predictive performances. HA.N. and N.M. analysed all results. HA.N. and N.M. wrote the manuscript with comments and feedback of all of the co-authors. All authors read and approved the manuscript.
Corresponding author
Ethics declarations
Competing interests
All authors declare no competing of interests.
Peer review
Peer review information
Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Nguyen, HA., Peleg, A.Y., Wisniewski, J.A. et al. AMR-GNN: a multi-representation graph neural network framework to enable genomic antimicrobial resistance prediction. Nat Commun (2026). https://doi.org/10.1038/s41467-026-69934-8
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-026-69934-8
