Discovering conserved epitopes of Monkeypox: Novel immunoinformatic and machine learning approaches

Abstract

The Monkeypox virus, an Orthopoxvirus with zoonotic origins, has been responsible for a growing number of human infections reminiscent of smallpox since May 2022, as reported by the World Health Organization. As of now, there are no established medical treatments for managing Monkeypox infections. In this study, we used machine learning to select conserved epitopes. Proteins were determined using Reverse Vaccinology and Gene Ontology subcellular localization, and their epitopes were predicted. NextClade was used to calculate the number of mutations in each amino acid position using 2433 Monkeypox sequences. The Unsupervised Nearest Neighbor machine learning algorithm and ideal matrix [0 0] were used to calculate the conservancy score of epitopes. Six proteins were determined for epitope prediction. Finally, 47 MHC-I epitopes, 5 MHC-II epitopes, and 10 Linear B cell epitopes were discovered. Our method can select epitopes for vaccine design to prevent viruses with accelerated evolution and high mutation rate.

Keywords

• Monkeypox
• Conserved epitopes
• Machine learning
• Immunoinformatic