AMR-GNN: a multi-representation graph neural network framework to enable genomic antimicrobial resistance prediction - PubMed
4 days ago
- #graph neural networks
- #genomic prediction
- #antimicrobial resistance
- AMR-GNN is a graph neural network framework designed for antimicrobial resistance (AMR) prediction from genomic data.
- It addresses challenges in AMR phenotype prediction by integrating multiple genomic representations.
- Tested on Pseudomonas aeruginosa, AMR-GNN shows potential in handling complex AMR mechanisms.
- The framework aims to enhance performance, mitigate clonal relationship influence, and identify biomarkers for explainability.
- Validation on a large dataset indicates broad applicability across diverse pathogen-drug combinations.
- The study highlights the importance of data-driven machine learning approaches in combating AMR.