Inter‐Verified Binary Classification Models: Unveiling Key Polymer Features Against Staphylococcus aureus
Carbon Innovation · 2026-06 · 1 (1) · 86-98
DOI: 10.1002/cin4.70003 ↗Kurz gefasst
Fachartikel in der Zeitschrift Carbon Innovation, online erschienen am 06.05.2026. Verfasst von Jooyoung Roh und 2 weiteren Autor:innen. Die Arbeit nennt 67 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Australian Research Council.
Abstract (Originalsprache)
Antimicrobial peptides (AMPs) are promising alternatives to antibiotics but require precise and costly sequence control for efficacy. Synthetic polymers composed of acrylate and acrylamide offer a simpler and more time‐efficient antibacterial strategy compared to AMPs. Although their activity against Gram‐negative bacteria has been widely studied, their potential against Gram‐positive bacteria remains relatively understudied. Here, decision tree, random forest and K‐nearest neighbors classification models were applied to a dataset of synthetic polymers targeting Gram‐positive bacteria Staphylococcus aureus to extensively investigate its modes of action and streamline the development of highly effective polymers that fulfilled the recommended characteristics. The predictive characteristics of an effective antibacterial polymer against S. aureus were (ranked in descending order of feature importance) the following: (1) DP limited to 40, (2) net‐charge limited to +30, (3) hydrophobic‐to‐cationic ratio 0.6 to 1, (4) hydrophobic composition between 0.35 and 0.45, (5) cationic composition between 0.45 and 0.55, (6) cLogP in between 0 and +1.5, and (7) the inclusion of aromatic rings. These key characteristics are attributed to the unique structure of Gram‐positive bacteria, where synthetic polymers are required to effectively diffuse through the moderately anionic thick cell wall to actively disrupt the plasmic membrane by relying on strong hydrophobic interactions.
Förderung
- Australian Research Council
Zitieren
APA
Roh, J., Boyer, C., & Kumar, P. (2026). Inter‐Verified Binary Classification Models: Unveiling Key Polymer Features Against Staphylococcus aureus. Carbon Innovation, 1(1), 86-98. https://doi.org/10.1002/cin4.70003
Vancouver
Roh J, Boyer C, Kumar P. Inter‐Verified Binary Classification Models: Unveiling Key Polymer Features Against Staphylococcus aureus. Carbon Innovation. 2026;1(1):86-98. doi:10.1002/cin4.70003
BibTeX
@article{Roh2026,
author = {Roh, Jooyoung and Boyer, Cyrille and Kumar, Priyank},
title = {Inter‐Verified Binary Classification Models: Unveiling Key Polymer Features Against Staphylococcus aureus},
journal = {Carbon Innovation},
year = {2026},
volume = {1},
number = {1},
pages = {86--98},
doi = {10.1002/cin4.70003}
}
EUVIMED