Machine learning, network pharmacology, and molecular dynamics reveal potent cyclopeptide inhibitors against dengue virus proteins

• Published in: Molecular diversity 2024-09
• Main Authors: Imam, Mohammed A, Alandijany, Thamir A, Felemban, Hashim R, Attar, Roba M, Faizo, Arwa A, Gattan, Hattan S, Dwivedi, Vivek Dhar, Azhar, Esam I
• Format: Article
• Language: English
• ISSN: 1381-1991; 1573-501X; 1573-501X
• DOI: 10.1007/s11030-024-10975-w

The dengue virus is a major global health hazard responsible for an estimated 390 million diseases yearly. This study focused on identifying cyclopeptide inhibitors for envelope structural proteins E, NS1, NS3, and NS5. Additionally, 5579 cyclopeptides were individually screened against the four target proteins using a machine learning-based quantitative structure-activity relationship model. Subsequently, the best 10 cyclopeptides from each protein were selected for molecular docking with their corresponding proteins. Moreover, the protein-peptide complexes with the highest affinity were subjected to a 100-ns molecular dynamics simulation. The protein-protein complexes exhibited superior structural stability and binding interactions. Based on the results of the MD simulation analyses, which included checking values for Root Mean Square Deviation, Root Mean Square Fluctuation, Principal Component Analysis (PCA), free energy landscapes, and energetic components, it was found that NS5-CP03714 complex is more stable and has stronger binding interactions than NS3-CP02054. PCA and free energy landscape plots have confirmed the higher conformational stability of NS5-CP03714. Analysis of the energetic components revealed that NS5-CP03714 (total binding energy = - 47.19 kcal/mol) exhibits more favorable interaction energies and overall binding energy compared to NS3-CP02054 (total binding energy = - 27.36 kcal/mol), suggesting a stronger and more stable formation of the complex. In addition, the drug-target network of two specific peptides (CP02950 and CP05582) and their associated target proteins were analyzed. This analysis revealed valuable information about their ability to target several proteins and their potential for broad-spectrum activity. Additional experimental investigations are necessary to validate these computational results and assess the efficacy of identified peptide inhibitors in biological systems.