Structure-aware graph learning links peptide geometry to pathogen-specific antimicrobial potency
. Capacity-matched ablations supported complementary contributions from predicted structure. Retrospective ranking recovered more known active peptides than the evaluated comparator, supporting further evaluation of GeoMIC for experimental candidate prioritization.
Authors
- César de la Fuente‐Núñez (ORCID: https://orcid.org/0000-0002-2005-5629)
- Jiangning Song (ORCID: https://orcid.org/0000-0001-8031-9086)
- 许舒婷
- Simon R. Corrie (ORCID: https://orcid.org/0000-0001-8029-1896)
- Shibo Kuang (ORCID: https://orcid.org/0000-0002-2969-9420)
- Anton Yariv Peleg (ORCID: https://orcid.org/0000-0002-2296-2126)
- Eric Tang (ORCID: https://orcid.org/0009-0008-7895-9250)
- Zehua Sun
- Xiaoyu Wang
Institutions
- The Alfred Hospital (AU)
- Monash University (AU)
- University of Pennsylvania (US)
Publication Details
- Journal
- Chemical Communications
- Published
- 2026-11-17
- DOI
- https://doi.org/10.1039/d6cc03538h
- Primary Topic
- Antimicrobial Peptides and Activities
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- University of Pennsylvania
- Monash University
- National Health and Medical Research Council