Comparison of deep learning models with simple method to assess the problem of antimicrobial peptides prediction
Molecular Informatics · 2024
DOI: 10.1002/minf.202200181 ↗Kurz gefasst
Vergleichsstudie in der Zeitschrift Molecular Informatics, 2024. Verfasst von M Y Lobanov und 6 weiteren Autor:innen. Beteiligte Einrichtung: Laboratory of Bioinformatics and Proteomics, Institute of Protein Research, Russian Academy of Sciences, 142290, Pushchino, Moscow Region, Russia. Hauptthemen laut MEDLINE: Deep Learning, Antimicrobial Peptides. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 25 Literaturangaben. Gefördert von: Russian science foundation.
Abstract (Originalsprache)
Antibiotic‐resistant strains are an emerging threat to public health. The usage of antimicrobial peptides (AMPs) is one of the promising approaches to solve this problem. For the development of new AMPs, it is necessary to have reliable prediction methods. Recently, deep learning approaches have been used to predict AMP. In this paper, we want to compare simple and complex methods for these purposes. We used the BERT transformer to create sequence embeddings and the multilayer perceptron (MLP) and light attention (LA) approaches for classification. One of them reached about 80 % accuracy and specificity in benchmark testing, which is on par with the best available methods. For comparison, we proposed a simple method using only the amino acid composition of proteins or peptides. This method has shown good results, at the level of the best methods. We have prepared a special server for predicting the ability of AMPs by amino acid composition: http://bioproteom.protres.ru/antimicrob/ .
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Laboratory of Bioinformatics and Proteomics, Institute of Protein Research, Russian Academy of Sciences, 142290, Pushchino, Moscow Region, Russia.
- Faculty of Applied math, MIREA - Russian Technological University, Moscow, 119454, Russia.
- Institute of Mathematical Problems of Biology branch of Keldysh Institute of Applied Mathematics, Russian Academy of Sciences, 142290, Pushchino, Russia.
- Laboratory of Structure and Function of Muscle Proteins, Institute of Theoretical and Experimental Biophysics, Russian Academy of Sciences, 142290, Pushchino, Moscow Region, Russia.
ORCID-Profile: O V Galzitskaya
Förderung
- Russian science foundation
Zitieren
APA
Lobanov, M. Y., Slizen, M. V., Dovidchenko, N. V., Panfilov, A. V., Surin, A. A., Likhachev, I. V., & Galzitskaya, O. V. (2024). Comparison of deep learning models with simple method to assess the problem of antimicrobial peptides prediction. Molecular Informatics. https://doi.org/10.1002/minf.202200181
Vancouver
Lobanov MY, Slizen MV, Dovidchenko NV, Panfilov AV, Surin AA, Likhachev IV, et al. Comparison of deep learning models with simple method to assess the problem of antimicrobial peptides prediction. Molecular Informatics. 2024. doi:10.1002/minf.202200181
BibTeX
@article{Lobanov2024,
author = {Lobanov, M Y and Slizen, M V and Dovidchenko, N V and Panfilov, A V and Surin, A A and Likhachev, I V and Galzitskaya, O V},
title = {Comparison of deep learning models with simple method to assess the problem of antimicrobial peptides prediction},
journal = {Molecular Informatics},
year = {2024},
doi = {10.1002/minf.202200181}
}
EUVIMED