The neuraminidase (NA) epitope from the Mexican AH1N1 influenza virus was identified by using sequences registered at the GenBank during the peak of a pandemic (from April 2009 to October 2010). First, NA protein sequences were submitted for multiple alignment analysis, and their three-dimensional models (3-D) were then built by using homology modeling. The most common sequence (denominated wild-type) and its mutants were submitted to linear and nonlinear epitope predictors, which included the major histocompatibility complex type II (MHC II) and B-cell peptides. The epitope prediction was in accordance with evolutionary behavior and some protein structural properties. The latter included a low NA mutation rate, NA 3-D surface exposure, and the presence of high hindrance side chain residues. After selecting the epitope, docking studies and molecular dynamics (MD) simulations were used to explore interactions between the epitope and MHC II. Afterward, several experimental assays were performed to validate the theoretical study by using antibodies from humans (infected by pandemic H1N1) and rabbits (epitope vaccination). The results show 119 complete sequences that were grouped into 28 protein sequences according to their identity (one wild-type and 27 representative mutants (1-5 mutations)). The predictors yielded several epitopes, with the best fit being the one located in the C-terminal region. Theoretical methods demonstrated that the selected epitope reached the P4, P6, P7, and P9 pockets of MHC II, whereas the experimental evidence indicates that the epitope is recognized by human antibodies and also by rabbit antibodies immunized with the peptide. © 2013 Springer Science+Business Media New York.