In Silico Prediction and Structural Mapping of Linear B-Cell Epitopes in Andes Virus Glycoprotein as Potential Vaccine Candidates

Abstract

Andes virus (ANDV), a pathogenic orthohantavirus associated with hantavirus cardiopulmonary syndrome, remains a public health concern because of its severe clinical outcomes and the absence of broadly approved vaccines for human use. The viral glycoprotein is a relevant antigenic target because of its surface localization and role in viral entry. This study aimed to identify and structurally map linear B-cell epitope candidates in the Andes virus glycoprotein using an immunoinformatics-based approach. The ANDV glycoprotein sequence was retrieved from UniProt, and linear B-cell epitopes were predicted using BepiPred 3.0 with a threshold of 0.030. Candidate peptides were further evaluated for antigenicity, allergenicity, and toxicity using VaxiJen v2.0, AllerTOP v2.1, and ToxinPred, respectively. The three-dimensional glycoprotein model was generated using AlphaFold, refined with GalaxyWEB, and validated using ERRAT and Ramachandran plot analysis. Five preliminary epitope candidates were identified, among which E1, E2, E3, and E4 met the antigenicity criterion. The refined structure showed favorable validation parameters, with an ERRAT score of 97.73 and 97.88% of residues in the favored Ramachandran region. Structural mapping indicated that E1 (EWRKKSDTTDTTNAASTTFE) and E2 (QTKTVNLRGT) were positioned on exposed surface regions, supporting their prioritization as the most promising candidates. E4 (PSTSSTDT) also showed potential, whereas E3 (HPRGEDHD), despite having the highest antigenicity score, appeared less structurally exposed. These findings provide preliminary computational evidence for surface-accessible linear B-cell epitope candidates in ANDV glycoprotein. However, experimental validation is required to confirm their immunological relevance.