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

A comparison between 2D and 3D descriptors in QSAR modeling based on bio‐active conformations

Malkeet Singh Bahia, Omer Kaspi, Meir Touitou, Idan Binayev, Seema Dhail, Jacob Spiegel iD, Netaly Khazanov, Abraham Yosipof, Hanoch Senderowitz iD

Molecular Informatics · 2023-04 · 42 (4)

DOI: 10.1002/minf.202200186 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Molecular Informatics, online erschienen am 17.02.2023. Verfasst von Malkeet Singh Bahia und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Chemistry, Bar-Ilan University, Ramat-Gan, 5290002, Israel. Hauptthemen laut MEDLINE: Quantitative Structure-Activity Relationship, Proteins. Die Arbeit wurde bisher 61-mal zitiert (Crossref) und nennt 69 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

QSAR models are widely and successfully used in many research areas. The success of such models highly depends on molecular descriptors typically classified as 1D, 2D, 3D, or 4D. While 3D information is likely important, e. g., for modeling ligand‐protein binding, previous comparisons between the performances of 2D and 3D descriptors were inconclusive. Yet in such comparisons the modeled ligands were not necessarily represented by their bioactive conformations. With this in mind, we mined the PDB for sets of protein‐ligand complexes sharing the same protein for which uniform activity data were reported. The results, totaling 461 structures spread across six series were compiled into a carefully curated, first of its kind dataset in which each ligand is represented by its bioactive conformation. Next, each set was characterized by 2D, 3D and 2D + 3D descriptors and modeled using three machine learning algorithms, namely, k ‐Nearest Neighbors, Random Forest and Lasso Regression. Models’ performances were evaluated on external test sets derived from the parent datasets either randomly or in a rational manner. We found that many more significant models were obtained when combining 2D and 3D descriptors. We attribute these improvements to the ability of 2D and 3D descriptors to code for different, yet complementary molecular properties.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Quantitative Structure-Activity RelationshipProteinsLigandsMolecular ConformationAlgorithms

Stichwörter der Autor:innen

2D vs. 3D descriptors comparison2D-descriptors3D-descriptorsbioactive conformationsmachine learning

Substanzen

LigandsProteins

Autor:innen und Institutionen

  • Department of Chemistry, Bar-Ilan University, Ramat-Gan, 5290002, Israel.
  • Current address: BCS program, UBC, Vancouver, Canada.
  • School of Cancer and Pharmaceutical Sciences, King's College London, London, 150 Stamford Street, SE1 9NH, United Kingdom.
  • Current address: Syneos Health Medical Communication Europe, London, UK.
  • Department of Information Systems, College of Law & Business, Ramat-Gan, P.O. Box 852, Bnei Brak, 5110801, Israel.

ORCID-Profile: Jacob Spiegel, Hanoch Senderowitz

Zitieren

APA

Bahia, M. S., Kaspi, O., Touitou, M., Binayev, I., Dhail, S., Spiegel, J., Khazanov, N., Yosipof, A., & Senderowitz, H. (2023). A comparison between 2D and 3D descriptors in QSAR modeling based on bio‐active conformations. Molecular Informatics, 42(4). https://doi.org/10.1002/minf.202200186

Vancouver

Bahia MS, Kaspi O, Touitou M, Binayev I, Dhail S, Spiegel J, et al. A comparison between 2D and 3D descriptors in QSAR modeling based on bio‐active conformations. Molecular Informatics. 2023;42(4). doi:10.1002/minf.202200186

BibTeX
@article{Bahia2023,
  author = {Bahia, Malkeet Singh and Kaspi, Omer and Touitou, Meir and Binayev, Idan and Dhail, Seema and Spiegel, Jacob and Khazanov, Netaly and Yosipof, Abraham and Senderowitz, Hanoch},
  title = {A comparison between 2D and 3D descriptors in QSAR modeling based on bio‐active conformations},
  journal = {Molecular Informatics},
  year = {2023},
  volume = {42},
  number = {4},
  doi = {10.1002/minf.202200186}
}
Kostenlos starten →