EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

PepFoundry: A Pipeline for Building Machine-Learning Ready Representations of Nonstandard Peptides Containing Cycles, Non-natural Residues, Polymer Units, and More

Daniel Garzon Otero iD, Omid Akbari, Aneesh Mandapati, Camille Bilodeau iD

Journal of Chemical Information and Modeling · 26.01.2026 · 66 (2) · 1264-1273

DOI: 10.1021/acs.jcim.5c02629 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Chemical Information and Modeling, online erschienen am 13.01.2026. Verfasst von Daniel Garzon Otero und 3 weiteren Autor:innen. Beteiligte Einrichtung: University of Virginia, Chemical Engineering Department, 385 McCormick Road, Charlottesville, Virginia 22903, United States. Hauptthemen laut MEDLINE: Peptides, Machine Learning, Software, Polymers. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Peptides featuring synthetic modifications, such as noncanonical amino acids, backbone modifications, cyclic structures, and polymer units have become central to modern drug design due to their enhanced stability and functional diversity. However, current machine learning (ML) approaches are restricted by challenges associated with transforming peptide sequences into atom-level representations, leading ML efforts to focus largely on datasets containing linear peptides comprised of standard residues. Here, we present PepFoundry, a Python package that handles peptide sequences beyond canonical amino acids and linear topologies by using SMILES strings in the CHUCKLES format. PepFoundry generates atom-mapped RDKit molecule objects, enabling the extraction of atom-level features, such as Morgan fingerprints and graph representations. We demonstrate its utility by processing a dataset of peptide sequences containing noncanonical amino acids and generating atomic level features for downstream property prediction. We show that atomic-level representations of peptides containing noncanonical amino acids consistently outperform sequence-level representations, regardless of model type. We additionally explore the representation of noncanonical peptides through latent space visualization and show that models with atomic-level information can effectively learn relationships between analogous sequences of l-peptides, d-peptides, and peptoids. This framework allows for the flexible incorporation of new amino acid chemistries, enabling existing ML methods to be straightforwardly applied to datasets of peptides containing nonstandard features. It also facilitates the rapid construction of customized peptide libraries and provides a scalable platform to accelerate ML-driven peptide discovery and optimization.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

PeptidesMachine LearningSoftwarePolymersAmino AcidsAmino Acid Sequence

Substanzen

PeptidesPolymersAmino Acids

Autor:innen und Institutionen

  • University of Virginia, Chemical Engineering Department, 385 McCormick Road, Charlottesville, Virginia 22903, United States.

ORCID-Profile: Daniel Garzon Otero, Camille Bilodeau

Zitieren

APA

Garzon Otero, D., Akbari, O., Mandapati, A., & Bilodeau, C. (2026). PepFoundry: A Pipeline for Building Machine-Learning Ready Representations of Nonstandard Peptides Containing Cycles, Non-natural Residues, Polymer Units, and More. Journal of Chemical Information and Modeling, 66(2), 1264-1273. https://doi.org/10.1021/acs.jcim.5c02629

Vancouver

Garzon Otero D, Akbari O, Mandapati A, Bilodeau C. PepFoundry: A Pipeline for Building Machine-Learning Ready Representations of Nonstandard Peptides Containing Cycles, Non-natural Residues, Polymer Units, and More. Journal of Chemical Information and Modeling. 2026;66(2):1264-1273. doi:10.1021/acs.jcim.5c02629

BibTeX
@article{GarzonOtero2026,
  author = {Garzon Otero, Daniel and Akbari, Omid and Mandapati, Aneesh and Bilodeau, Camille},
  title = {PepFoundry: A Pipeline for Building Machine-Learning Ready Representations of Nonstandard Peptides Containing Cycles, Non-natural Residues, Polymer Units, and More},
  journal = {Journal of Chemical Information and Modeling},
  year = {2026},
  volume = {66},
  number = {2},
  pages = {1264--1273},
  doi = {10.1021/acs.jcim.5c02629}
}
Kostenlos starten →