Antimicrobial Resistance Predictor

Data Downloading

Download curated datasets, models, and code for reproducibility and further analysis.

The source code for model training, prediction, and analysis is available on GitHub under the MIT License.
Curated genome assemblies and manually checked antibiograms are available on Zenodo. This repository includes:
- Machine learning models with performance metrics and SHAP values: models_shap_values.zip
- Encoded k-mer datasets for training: kmer3_data.zip, kmer4_data.zip, kmer5_data.zip
- Assemblies and antibiograms of 36 metagenomic blood culture samples: bloodcultures_assemblies_antibiograms.zip
- K-mer profiles for resistance phenotype prediction: kmers_blood_cultures.zip
- Filtered high-quality assemblies and antibiograms: filtered_assemblies.zip, qced_antibiograms.zip
Raw metagenomic sequencing data (FASTQ files) from 40 patients with bloodstream infections are deposited in the NCBI Sequence Read Archive (SRA) under BioProject accession PRJNA1290626. This includes 52 Gbases of multispecies data across 40 SRA experiments and 40 BioSamples, collected at the University Hospital of Larissa (registration date: 12-Jul-2025).

Cite This Work

If you use this tool or data, please cite our publication

Machine Learning-based Prediction of Antibiotic Resistance from Genomic Data

Anargyros Skoulakis, Konstantinos Daniilidis, Stefanos Digenis, Christos-Georgios Gkountinoudis, Efthimia Petinaki, Artemis G. Hatzigeorgiou

Research Square2026DOI: 10.21203/rs.3.rs-7190203/v1

Export Citation

BibTeX

@article{skoulakis2026amr,
  title = {Machine Learning-based Prediction of Antibiotic Resistance from Genomic Data},
  author = {Skoulakis, Anargyros and Daniilidis, Konstantinos and Digenis, Stefanos and Gkountinoudis, Christos-Georgios and Petinaki, Efthimia and Hatzigeorgiou, Artemis G.},
  journal = {Research Square},
  year = {2026},
  doi = {10.21203/rs.3.rs-7190203/v1},
  url = {https://www.researchsquare.com/article/rs-7190203/v1}
}