EUVIMEDEUVIMED entdecken
ÜbersichtsarbeitOpen Access · CC BY 4.0

Comparative Studies on Resampling Techniques in Machine Learning and Deep Learning Models for Drug-Target Interaction Prediction

Azwaar Khan Azlim Khan, Nurul Hashimah Ahamed Hassain Malim

Molecules · 2023

DOI: 10.3390/molecules28041663 ↗

Kurz gefasst

Übersichtsarbeit in der Zeitschrift Molecules, 2023. Verfasst von Azwaar Khan Azlim Khan und 1 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Sciences, Universiti Sains Malaysia, Pulau Pinang 11800, Malaysia. Hauptthemen laut MEDLINE: Deep Learning. Die Arbeit wurde bisher 35-mal zitiert (Crossref) und nennt 62 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Higher Education.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The prediction of drug-target interactions (DTIs) is a vital step in drug discovery. The success of machine learning and deep learning methods in accurately predicting DTIs plays a huge role in drug discovery. However, when dealing with learning algorithms, the datasets used are usually highly dimensional and extremely imbalanced. To solve this issue, the dataset must be resampled accordingly. In this paper, we have compared several data resampling techniques to overcome class imbalance in machine learning methods as well as to study the effectiveness of deep learning methods in overcoming class imbalance in DTI prediction in terms of binary classification using ten (10) cancer-related activity classes from BindingDB. It is found that the use of Random Undersampling (RUS) in predicting DTIs severely affects the performance of a model, especially when the dataset is highly imbalanced, thus, rendering RUS unreliable. It is also found that SVM-SMOTE can be used as a go-to resampling method when paired with the Random Forest and Gaussian Naïve Bayes classifiers, whereby a high F1 score is recorded for all activity classes that are severely and moderately imbalanced. Additionally, the deep learning method called Multilayer Perceptron recorded high F1 scores for all activity classes even when no resampling method was applied.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningBayes TheoremMachine LearningNeural Networks, ComputerAlgorithms

Stichwörter der Autor:innen

SMOTEclass imbalancedata resamplingdeep learningdrug-target interactionmachine learning

Autor:innen und Institutionen

  • School of Computer Sciences, Universiti Sains Malaysia, Pulau Pinang 11800, Malaysia.

Förderung

  • Ministry of Higher Education

Zitieren

APA

Azlim Khan, A. K., & Ahamed Hassain Malim, N. H. (2023). Comparative Studies on Resampling Techniques in Machine Learning and Deep Learning Models for Drug-Target Interaction Prediction. Molecules. https://doi.org/10.3390/molecules28041663

Vancouver

Azlim Khan AK, Ahamed Hassain Malim NH. Comparative Studies on Resampling Techniques in Machine Learning and Deep Learning Models for Drug-Target Interaction Prediction. Molecules. 2023. doi:10.3390/molecules28041663

BibTeX
@article{AzlimKhan2023,
  author = {Azlim Khan, Azwaar Khan and Ahamed Hassain Malim, Nurul Hashimah},
  title = {Comparative Studies on Resampling Techniques in Machine Learning and Deep Learning Models for Drug-Target Interaction Prediction},
  journal = {Molecules},
  year = {2023},
  doi = {10.3390/molecules28041663}
}
Kostenlos starten →