EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

BindingSite-AugmentedDTA: enabling a next-generation pipeline for interpretable prediction models in drug repurposing

Niloofar Yousefi, Mehdi Yazdani-Jahromi, Aida Tayebi, Elayaraja Kolanthai, Craig J Neal, Tanumoy Banerjee, Agnivo Gosai, Ganesh Balasubramanian, Sudipta Seal, Ozlem Ozmen Garibay

Briefings in Bioinformatics · 19.05.2023 · 24 (3)

DOI: 10.1093/bib/bbad136 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, online erschienen am 24.04.2023. Verfasst von Niloofar Yousefi und 9 weiteren Autor:innen. Beteiligte Einrichtung: Industrial Engineering and Management Systems, University of Central Florida, 32816, 4000 Central Florida Blvd., Orlando, FL, USA. Hauptthemen laut MEDLINE: Drug Repositioning, Algorithms. Die Arbeit wurde bisher 29-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

While research into drug–target interaction (DTI) prediction is fairly mature, generalizability and interpretability are not always addressed in the existing works in this field. In this paper, we propose a deep learning (DL)-based framework, called BindingSite-AugmentedDTA, which improves drug–target affinity (DTA) predictions by reducing the search space of potential-binding sites of the protein, thus making the binding affinity prediction more efficient and accurate. Our BindingSite-AugmentedDTA is highly generalizable as it can be integrated with any DL-based regression model, while it significantly improves their prediction performance. Also, unlike many existing models, our model is highly interpretable due to its architecture and self-attention mechanism, which can provide a deeper understanding of its underlying prediction mechanism by mapping attention weights back to protein-binding sites. The computational results confirm that our framework can enhance the prediction performance of seven state-of-the-art DTA prediction algorithms in terms of four widely used evaluation metrics, including concordance index, mean squared error, modified squared correlation coefficient ($r^2_m$) and the area under the precision curve. We also contribute to three benchmark drug–traget interaction datasets by including additional information on 3D structure of all proteins contained in those datasets, which include the two most commonly used datasets, namely Kiba and Davis, as well as the data from IDG-DREAM drug-kinase binding prediction challenge. Furthermore, we experimentally validate the practical potential of our proposed framework through in-lab experiments. The relatively high agreement between computationally predicted and experimentally observed binding interactions supports the potential of our framework as the next-generation pipeline for prediction models in drug repurposing.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Drug RepositioningAlgorithmsDrug DevelopmentProteinsBinding Sites

Stichwörter der Autor:innen

DTA databaseDTA softwareSARS-CoV-2binding sitesdeep learningdrug–target affinitydrug–target interactionmachine learning

Substanzen

Proteins

Autor:innen und Institutionen

  • Industrial Engineering and Management Systems, University of Central Florida, 32816, 4000 Central Florida Blvd., Orlando, FL, USA.
  • Computer Science, University of Central Florida, 32816, 4000 Central Florida Blvd., Orlando, FL, USA.
  • College of Medicine, Bionix Cluster, University of Central Florida, 4000 Central Florida Blvd., Orlando 32816, FL, USA.
  • Department of Mechanical Engineering and Mechanics, Lehigh University, Bethlehem 18015, PA, USA.
  • Independent Researcher, Corning, NY, USA.
  • Advanced Materials Processing and Analysis Center, Department of Materials Science and Engineering, University of Central Florida, 4000 Central Florida Blvd., Orlando 32816, FL, USA.

Zitieren

APA

Yousefi, N., Yazdani-Jahromi, M., Tayebi, A., Kolanthai, E., Neal, C. J., Banerjee, T., Gosai, A., Balasubramanian, G., Seal, S., & Ozmen Garibay, O. (2023). BindingSite-AugmentedDTA: enabling a next-generation pipeline for interpretable prediction models in drug repurposing. Briefings in Bioinformatics, 24(3). https://doi.org/10.1093/bib/bbad136

Vancouver

Yousefi N, Yazdani-Jahromi M, Tayebi A, Kolanthai E, Neal CJ, Banerjee T, et al. BindingSite-AugmentedDTA: enabling a next-generation pipeline for interpretable prediction models in drug repurposing. Briefings in Bioinformatics. 2023;24(3). doi:10.1093/bib/bbad136

BibTeX
@article{Yousefi2023,
  author = {Yousefi, Niloofar and Yazdani-Jahromi, Mehdi and Tayebi, Aida and Kolanthai, Elayaraja and Neal, Craig J and Banerjee, Tanumoy and Gosai, Agnivo and Balasubramanian, Ganesh and Seal, Sudipta and Ozmen Garibay, Ozlem},
  title = {BindingSite-AugmentedDTA: enabling a next-generation pipeline for interpretable prediction models in drug repurposing},
  journal = {Briefings in Bioinformatics},
  year = {2023},
  volume = {24},
  number = {3},
  doi = {10.1093/bib/bbad136}
}
Kostenlos starten →