To enhance utility of the linear epitope mapping (Pepscan) technique for assay of humoral responses linked to vaccination, two modifications were tested. First, peptides were incubated with serum contained in baths rather than individual wells. Second, a rigorous statistical model was developed to determine which peptide/antibody-binding interactions were significant. The modifications increased the ability to detect signal in these experiments by 15- to 45-fold. These two modifications were applied to linear epitope mapping of HIV seropositive volunteers under treatment with recombinant HIV gp160 and also to rabbits immunized with the same product. Changes in fine specificity of response were observed in animal models and human vaccine recipients over the course of an immunization series with this antigen.