The application of a recently developed technique of conformational memories (Guarnieri, F.; Wilson, S. R. J. Comput. Chem. 1995, 16, 5, 648) to the study of peptide structure is illustrated with an analysis of the conformational populations of the decapeptide gonadotropin-releasing hormone (GnRH) and the mutant peptide Lys8-GnRH. The conformational space of the peptides is explored fully with multiple Monte Carlo simulated annealing (MC/SA) random walks using the Amber* force field and the generalized Born/surface area (GB/SA) continuum solvation model of water as implemented in the Macromodel molecular modeling package. The collective histories of all the random walks are transformed into mean field dihedral distribution functions called “conformational memories” (35 conformational memories for GnRH, one for each torsion angle). Conformational families of the peptides at 310 K are obtained from these results with the use of a biased sampling technique which explores only the areas found to be populated in the conformational memories. A large family of GnRH structures that comprises approximately 70% of the population is identified by clustering conformations according to a pairwise root mean square deviation criterion. Members of this family of conformations share a distinctive β-type turn in the backbone that involves residues 5−8. This conformation is shown to correspond to the preferred geometry of a structurally constrained analog of GnRH that binds to the GnRH receptor with high affinity. In contrast, GnRH analogs such as Lys8-GnRH, for which we find that the major conformational family exhibits an extended backbone, seem to belong to the group of low-affinity ligands. The results suggest a correlation between high affinity at the GnRH receptor and the ability of the peptides to have a large population of conformations that exhibit the characteristic secondary structure of a β-type bend. Thus, the method of conformational memories is shown to provide a powerful and reliable tool for the computational exploration of structure−function relations of flexible peptides.