Immune system response requires T-cell activation. On the molecular level, peptide-specific signals delivered through the engagement of a T-cell receptor (TCR) on the cell surface with the peptide-major histocompatibility complex (pMHC) on the surface of an antigen-presenting cell (APC). Although additional co-stimulatory signals are delivered through the interactions of accessory molecules on the T-cell surface with complexes on the surface of the APC, TCR activation depends critically on the peptide-mediated binding affinity of the TCR with the MHC. The alteration of a single amino acid residue of the peptide, for example, can produce a dramatic change from no response at all to a strong response. Surface plasmon resonance studies of TCR/pMHC interactions show that while the range of binding affinities of stimulatory pMHC ligands is low compared to that for antibody-antigen interactions, they need to be sufficiently high to induce T-cell activation. This suggests that important local interactions need to be identified in an otherwise ``noisy" background. In this talk, we describe a novel method based on the Potential Distribution Theorem and large-scale molecular dynamics simulations for computing relative TCR/pMHC binding affinities as a function of the amino acid sequence of the peptide. Results are shown for the human T-cell lymphotropic virus Tax peptide (LLFGYPVYV) and a single amino acid mutant (P6A) bound to the A6/HLA-A2 T-cell/MHC complex. These results are discussed in the context of the development of complementary high-throughput pMHC microarrays and bioinformatic databases to screen for peptide antigens.