Investigating molecular descriptors in cell-penetrating peptides prediction with deep learning: Employing N, O, and hydrophobicity according to the Eisenberg scale

• Published in: PloS one Vol. 19; no. 6; p. e0305253
• Main Authors: Seixas Feio, Juliana Auzier, de Oliveira, Ewerton Cristhian Lima, de Sales, Junior, Claudomiro de Souza, da Costa, Kauê Santana, E Lima, Anderson Henrique Lima
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 13.06.2024
• Subjects: Accuracy; Algorithms; Amino acids; Analysis; Artificial intelligence; Artificial neural networks; Biology and Life Sciences; Cell-Penetrating Peptides - chemistry; Cell-Penetrating Peptides - metabolism; Classification; Computer and Information Sciences; Datasets; Deep Learning; Drug delivery; Engineering and Technology; Hydrogen bonds; Hydrophobic and Hydrophilic Interactions; Hydrophobicity; Innovations; Investigations; Lipid bilayers; Lipids; Machine learning; Membrane lipids; Membrane permeability; Membranes; Neural networks; Neural Networks, Computer; Nitrogen; Nitrogen - chemistry; Oxygen; Oxygen - chemistry; Oxygen - metabolism; Peptides; Permeability; Physical Sciences; Proteins; Research and Analysis Methods; Support vector machines; Wavelet transforms
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0305253

Cell-penetrating peptides comprise a group of molecules that can naturally cross the lipid bilayer membrane that protects cells, sharing physicochemical and structural properties, and having several pharmaceutical applications, particularly in drug delivery. Investigations of molecular descriptors have provided not only an improvement in the performance of classifiers but also less computational complexity and an enhanced understanding of membrane permeability. Furthermore, the employment of new technologies, such as the construction of deep learning models using overfitting treatment, promotes advantages in tackling this problem. In this study, the descriptors nitrogen, oxygen, and hydrophobicity on the Eisenberg scale were investigated, using the proposed ConvBoost-CPP composed of an improved convolutional neural network with overfitting treatment and an XGBoost model with adjusted hyperparameters. The results revealed favorable to the use of ConvBoost-CPP, having as input nitrogen, oxygen, and hydrophobicity together with ten other descriptors previously investigated in this research line, showing an increase in accuracy from 88% to 91.2% in cross-validation and 82.6% to 91.3% in independent test.