Interpreting Graph Neural Networks for Antibacterial Drug Repurposing against Acinetobacter baumannii
- 1University of Padova
Abstract
Antimicrobial resistance in Acinetobacter baumannii, whose carbapenem-resistant strains are classified by the World Health Organization as a critical priority, has motivated machine-learning screens for new antibacterials, but translating predictions into chemical hypotheses requires evaluating the molecular features on which a model relies. A graph neural network ensemble achieves competitive predictive performance against a descriptor-augmented reference model on the same antibacterial screen while using only atom features and connectivity, allowing interpretation to focus on the molecular graph without a separate input of handcrafted descriptors. The atom-level explanations identify features supporting the model's activity predictions under perturbation and restoration, while comparisons with equally sized perturbations within each molecule distinguish piperazine as the most consistently supported recurring contributor to predicted activity. Prospective screening of the Drug Repurposing Hub followed by developability and safety assessment yields a ranked shortlist for experimental follow-up and leads to testing of CE3F4, which reduced A. baumannii growth by 57% at 0.25 mM in a preliminary assay with a non-monotonic concentration--response profile.
Related
- Design of a Human Defensin 5-Derived Linear Peptide with Potent Antibacterial Efficacy Against Carbapenem-Resistant Acinetobacter baumannii
- Metabolic acidification–linked outer membrane vesiculation in Acinetobacter baumannii
- Receptor-guided phage cocktails constrain Acinetobacter baumannii resistance and improve therapy
- Polystyrene Microplastics Accelerate Antibiotic Resistance Evolution and Exacerbate Pathogenicity in Acinetobacter Baumannii
- Beyond the Axenic Culture Filtrate Secretome: BONCAT and conventional proteomics together reveal the Infection-associated secreted protein repertoire of Acinetobacter baumannii in the human lung Epithelial cell line A549
BibTeX
@misc{romano2026interpreting,
title = {{Interpreting Graph Neural Networks for Antibacterial Drug Repurposing against Acinetobacter baumannii}},
author = {S. Romano and M. Grassia and D. Righelli and P. Lio and G. Mangioni and C. Taccioli},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.26.761825},
url = {https://www.biorxiv.org/content/10.64898/2026.10.26.761825v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Interpreting Graph Neural Networks for Antibacterial Drug Repurposing against Acinetobacter baumannii
AU - S. Romano
AU - M. Grassia
AU - D. Righelli
AU - P. Lio
AU - G. Mangioni
AU - C. Taccioli
PY - 2026
DA - 2026/10/26
PB - bioRxiv
DO - 10.64898/2026.10.26.761825
UR - https://www.biorxiv.org/content/10.64898/2026.10.26.761825v1
AB - Antimicrobial resistance in Acinetobacter baumannii, whose carbapenem-resistant strains are classified by the World Health Organization as a critical priority, has motivated machine-learning screens for new antibacterials, but translating predictions into chemical hypotheses requires evaluating the molecular features on which a model relies. A graph neural network ensemble achieves competitive predictive performance against a descriptor-augmented reference model on the same antibacterial screen while using only atom features and connectivity, allowing interpretation to focus on the molecular graph without a separate input of handcrafted descriptors. The atom-level explanations identify features supporting the model's activity predictions under perturbation and restoration, while comparisons with equally sized perturbations within each molecule distinguish piperazine as the most consistently supported recurring contributor to predicted activity. Prospective screening of the Drug Repurposing Hub followed by developability and safety assessment yields a ranked shortlist for experimental follow-up and leads to testing of CE3F4, which reduced A. baumannii growth by 57% at 0.25 mM in a preliminary assay with a non-monotonic concentration--response profile.
ER -