ESM-BBB-Pred: a fine-tuned ESM 2.0 and deep neural networks for the identification of blood–brain barrier peptides
Briefings in Bioinformatics · 2024
DOI: 10.1093/bib/bbaf066 ↗Kurz gefasst
Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2024. Verfasst von Ansar Naseem und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Software Engineering, Superior University, 17 KM Raiwind Road Lahore, Punjab 55150, Pakistan. Hauptthemen laut MEDLINE: Blood-Brain Barrier, Peptides, Neural Networks, Computer, Computational Biology. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Blood–brain barrier peptides (BBBP) could significantly improve the delivery of drugs to the brain, paving the way for new treatments for central nervous system (CNS) disorders. The primary challenge in treating CNS disorders lies in the difficulty pharmaceutical agent’s face in crossing the BBB. Almost 98% of small molecule drugs and nearly all large molecule drugs fail to penetrate the BBB effectively. Thus, identifying these peptides is vital for advancements in healthcare. This study introduces an enhanced intelligent computational model called BBB-PEP- Evolutionary Scale Modeling (ESM), designed to identify BBBP. The relative positions, reverse position and statistical moment-based features have been utilized on the existing benchmark dataset. For classification purpose, six deep classifiers such as fully connected networks, convolutional neural network, simple recurrent neural networks, long short-term memory (LSTM), bidirectional LSTM, and gated recurrent unit have been utilized. In addition to harnessing the effectiveness of the pre-trained model, a protein language model ESM 2.0 has been fine-tuned on a benchmark dataset for BBBP classification. Three tests such as self-consistency, independent set testing, and five-fold cross-validation have been utilized for evaluation purposes with evaluation metrics includes accuracy, specificity, sensitivity, and Matthews correlation coefficient. The fine-tuned model ESM 2.0 has shown superior results as compared to employed classifiers and surpasses the existing benchmark studies. This system will support future research and the scientific community in the computational identification of BBBP.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Department of Software Engineering, Superior University, 17 KM Raiwind Road Lahore, Punjab 55150, Pakistan.
- Department of Cybersecurity, College of Computer, Qassim University, Buraydah, Saudi Arabia.
- Department of Computer Engineering, College of Computer, Qassim University, Buraydah, Saudi Arabia.
- Department of Computer Science, School of Systems and Technology, University of Management and Technology, Lahore, Pakistan.
ORCID-Profile: Tamim Alkhalifah, Yaser Daanial Khan
Zitieren
APA
Naseem, A., Alturise, F., Alkhalifah, T., & Khan, Y. D. (2024). ESM-BBB-Pred: a fine-tuned ESM 2.0 and deep neural networks for the identification of blood–brain barrier peptides. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbaf066
Vancouver
Naseem A, Alturise F, Alkhalifah T, Khan YD. ESM-BBB-Pred: a fine-tuned ESM 2.0 and deep neural networks for the identification of blood–brain barrier peptides. Briefings in Bioinformatics. 2024. doi:10.1093/bib/bbaf066
BibTeX
@article{Naseem2024,
author = {Naseem, Ansar and Alturise, Fahad and Alkhalifah, Tamim and Khan, Yaser Daanial},
title = {ESM-BBB-Pred: a fine-tuned ESM 2.0 and deep neural networks for the identification of blood–brain barrier peptides},
journal = {Briefings in Bioinformatics},
year = {2024},
doi = {10.1093/bib/bbaf066}
}
EUVIMED