A Computational Workflow Integrating Explainable AI and Database Mining for the Prioritization of Antimicrobial Natural Products
5 affiliations
- 1Department of Surgical and Medical Sciences, Magna Graecia University of Catanzaro, Viale Europa, 88100 Catanzaro, Italy
- 2Department of Computer Engineering, Modeling, Electronics and Systems (DIMES), University of Calabria, Via P. Bucci, 87036 Rende, Italy
- 3Department of Public Health and Infectious Diseases, University Hospital Policlinico Umberto I, Sapienza University of Rome, 00185 Rome, Italy
- 4Department of Biomedical Sciences, University of Sassari, 07100 Sassari, Italy
- 5Unit of Medical Statistics and Molecular Epidemiology, Università Campus Bio-Medico di Roma, 00128 Rome, Italy
Abstract
The escalating threat of antimicrobial resistance demands the discovery of antibacterial agents with unprecedented chemical scaffolds. Natural products have historically been the most prolific source of antibiotics, yet the vast majority of catalogued natural products remain untested for antimicrobial activity. Here we present Natural-Product Graph–language, Uncertainty-Aware Relational Discovery (NP-GUARD), a modular computational framework that reframes antimicrobial natural-product discovery as an uncertainty-aware, relational, multi-objective learning problem. NP-GUARD integrates six core components: provenance-tracked data curation, multi-view molecular representation, pathogen-conditioned multi-task prediction, explicit uncertainty quantification, stability-aware explainability, and conformal-Pareto prioritization, all consolidated within a queryable knowledge base. Of these components, the present study implements and benchmarks the data-curation, predictive, uncertainty, explainability and prioritization layers, including a first pathogen-conditioned masked multi-task graph model, while the full multi-view encoder and the knowledge base are specified for subsequent work. We report a proof-of-concept evaluation on public screening data for Staphylococcus aureus (38,995 compounds), Escherichia coli (2291) and Pseudomonas aeruginosa (29,986), together with human cytotoxicity data and a 406,959-compound screening library from COCONUT. Under repeated scaffold-aware splitting, baseline models achieve strong discrimination (AUROC 0.78–0.93) but reveal substantial split-to-split variance in minority-class performance; class-conditional conformal prediction attains near-nominal per-class coverage; and feature attributions show only moderate stability across independently trained models (top-20 Jaccard 0.54), underscoring a critical caveat for explanation-driven discovery. Screening the COCONUT library validates the uncertainty-aware activity ranking, achieving an 87-fold enrichment of known antibiotics in the top 100 candidates (128-fold within the fully characterized named-compound stratum), although analogue-exclusion analysis shows this signal is dominated by rediscovery of training-like chemistry. The novelty-seeking Pareto layer yields a structurally diverse 90-compound shortlist spanning 72 scaffolds, deliberately steered toward unexplored chemotypes; we emphasize that the shortlist constitutes prioritized hypotheses for experimental testing, not validated actives. NP-GUARD provides a transparent, reproducible and decision-oriented foundation for prioritizing natural products for experimental validation.
Related
- Computational design and prioritization of database-absent small-molecule GLP-1 receptor agonist candidates from public bioactivity data
- Old Drugs, New Targets: Multi-parameter computational prioritization of FDA‑approved, brain‑penetrant drugs as isoform‑selective GSK‑3β inhibitors for Alzheimer’s disease
- Workflow design as a research variable in computational materials discovery
- A Simulation-Driven Digital Workflow Integrating Bayesian Optimization for Early-Stage Process Design: A Methyl Acetate Production Case Study
- AGAPE (Computational G-Quadruplex Stabilization Prediction): The First Machine Learning Workflow for G-Quadruplex Stabilization Prediction
BibTeX
@article{guzzi2026computational,
title = {{A Computational Workflow Integrating Explainable AI and Database Mining for the Prioritization of Antimicrobial Natural Products}},
author = {Pietro Hiram Guzzi and Annamaria Defilippo and Pierangelo Veltri and Giancarlo Ceccarelli and Fabio Scarpa and Massimo Ciccozzi and Francesco Branda},
journal = {Molecules},
year = {2026},
month = oct,
doi = {10.3390/molecules31203683},
url = {https://www.mdpi.com/1420-3049/31/20/3683},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - A Computational Workflow Integrating Explainable AI and Database Mining for the Prioritization of Antimicrobial Natural Products
AU - Pietro Hiram Guzzi
AU - Annamaria Defilippo
AU - Pierangelo Veltri
AU - Giancarlo Ceccarelli
AU - Fabio Scarpa
AU - Massimo Ciccozzi
AU - Francesco Branda
PY - 2026
DA - 2026/10/01
JO - Molecules
DO - 10.3390/molecules31203683
UR - https://www.mdpi.com/1420-3049/31/20/3683
AB - The escalating threat of antimicrobial resistance demands the discovery of antibacterial agents with unprecedented chemical scaffolds. Natural products have historically been the most prolific source of antibiotics, yet the vast majority of catalogued natural products remain untested for antimicrobial activity. Here we present Natural-Product Graph–language, Uncertainty-Aware Relational Discovery (NP-GUARD), a modular computational framework that reframes antimicrobial natural-product discovery as an uncertainty-aware, relational, multi-objective learning problem. NP-GUARD integrates six core components: provenance-tracked data curation, multi-view molecular representation, pathogen-conditioned multi-task prediction, explicit uncertainty quantification, stability-aware explainability, and conformal-Pareto prioritization, all consolidated within a queryable knowledge base. Of these components, the present study implements and benchmarks the data-curation, predictive, uncertainty, explainability and prioritization layers, including a first pathogen-conditioned masked multi-task graph model, while the full multi-view encoder and the knowledge base are specified for subsequent work. We report a proof-of-concept evaluation on public screening data for Staphylococcus aureus (38,995 compounds), Escherichia coli (2291) and Pseudomonas aeruginosa (29,986), together with human cytotoxicity data and a 406,959-compound screening library from COCONUT. Under repeated scaffold-aware splitting, baseline models achieve strong discrimination (AUROC 0.78–0.93) but reveal substantial split-to-split variance in minority-class performance; class-conditional conformal prediction attains near-nominal per-class coverage; and feature attributions show only moderate stability across independently trained models (top-20 Jaccard 0.54), underscoring a critical caveat for explanation-driven discovery. Screening the COCONUT library validates the uncertainty-aware activity ranking, achieving an 87-fold enrichment of known antibiotics in the top 100 candidates (128-fold within the fully characterized named-compound stratum), although analogue-exclusion analysis shows this signal is dominated by rediscovery of training-like chemistry. The novelty-seeking Pareto layer yields a structurally diverse 90-compound shortlist spanning 72 scaffolds, deliberately steered toward unexplored chemotypes; we emphasize that the shortlist constitutes prioritized hypotheses for experimental testing, not validated actives. NP-GUARD provides a transparent, reproducible and decision-oriented foundation for prioritizing natural products for experimental validation.
SN - 1420-3049
ER -