EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Circumventing the Synthesizability Problem in Generative Molecular Design

Jesse A Weller iD, Jinsen Li, Yibei Jiang, Remo Rohs iD

Journal of Chemical Information and Modeling · 27.07.2026 · 66 (14) · 8295-8303

DOI: 10.1021/acs.jcim.6c00998 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Chemical Information and Modeling, online erschienen am 09.07.2026. Verfasst von Jesse A Weller und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Quantitative and Computational Biology, University of Southern California, Los Angeles, California90089, United States. Hauptthemen laut MEDLINE: Drug Design. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: NIGMS NIH HHS, Office of Research and Innovation, University of Southern California, National Institutes of Health (NIH).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Generative structure-based drug design (SBDD) models have shown great promise for accelerating our ability to discover novel drug candidates. However, these models have been criticized for producing compounds that are not very synthesizable and therefore not practically applicable to drug design. In this work, we propose a way to circumvent the synthesizability issue by introducing a model-guided virtual screening (MGVS) pipeline that pairs SBDD models with efficient chemical similarity search methods to identify synthesizable analogues of generated compounds in existing ultralarge compound databases. Using this approach, we demonstrate that synthesizable analogues of generated compounds with equivalent or better docking scores and similar predicted binding poses can be reliably identified across a wide range of protein targets. We find that MGVS outperforms standard virtual ligand screening (VLS), consistently yielding at least a 25x improvement in docking-based screening efficiency across three different SBDD models. As drug-like chemical spaces continue to grow and standard VLS methods focused on exhaustive screening become increasingly impractical, approaches such as MGVS that effectively narrow the search space will become critical for advancing drug discovery.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Drug DesignLigandsMolecular Docking SimulationGenerative Artificial IntelligenceProteins

Substanzen

LigandsProteins

Autor:innen und Institutionen

  • Department of Quantitative and Computational Biology, University of Southern California, Los Angeles, California90089, United States.
  • Department of Physics & Astronomy, University of Southern California, Los Angeles, California90089, United States.
  • Department of Chemistry, University of Southern California, Los Angeles, California90089, United States.
  • Thomas Lord Department of Computer Science, University of Southern California, Los Angeles, California90089, United States.
  • Division of Medical Oncology, Department of Medicine, University of Southern California, Los Angeles, California90033, United States.
  • Alfred E. Mann Department of Biomedical Engineering, University of Southern California, Los Angeles, California90089, United States.

ORCID-Profile: Jesse A Weller, Remo Rohs

Förderung

  • NIGMS NIH HHS
  • Office of Research and Innovation, University of Southern California
  • National Institutes of Health (NIH)

Zitieren

APA

Weller, J. A., Li, J., Jiang, Y., & Rohs, R. (2026). Circumventing the Synthesizability Problem in Generative Molecular Design. Journal of Chemical Information and Modeling, 66(14), 8295-8303. https://doi.org/10.1021/acs.jcim.6c00998

Vancouver

Weller JA, Li J, Jiang Y, Rohs R. Circumventing the Synthesizability Problem in Generative Molecular Design. Journal of Chemical Information and Modeling. 2026;66(14):8295-8303. doi:10.1021/acs.jcim.6c00998

BibTeX
@article{Weller2026,
  author = {Weller, Jesse A and Li, Jinsen and Jiang, Yibei and Rohs, Remo},
  title = {Circumventing the Synthesizability Problem in Generative Molecular Design},
  journal = {Journal of Chemical Information and Modeling},
  year = {2026},
  volume = {66},
  number = {14},
  pages = {8295--8303},
  doi = {10.1021/acs.jcim.6c00998}
}
Kostenlos starten →