Unravelling the mystery of protein folding has been a primary goal of the biological sciences since Anfinsen's seminal experiments demonstrated the connection between a protein's amino acid sequence and its active three-dimensional conformation.[1] Since then, there has been steady progress in the experimental, [2] theoretical, [3] analytical [4] and predictive [5] aspects of the protein folding problem. However, the matter is far from resolved, [6] remaining one of the great open problems of our time. [7] A complete understanding of the forces acting on individual amino acids during folding would undoubtedly enhance our understanding of this fundamental process.One of the more pivotal amino acids for protein folding and three-dimensional structure is glycine: H 2 NCH 2 COOH. Lacking a side chain, this residue is capable of adopting many conformations that are sterically challenging for amino acids with larger side chains. Theoretical investigations of glycine conformations have focused on isolated glycine molecules in the gas phase, [8] with most reports maintaining that hydrogen bonding between the amine and carboxylic acid moieties is responsible for the preferred gas-phase planar geometries.[8a, d-h, j] The conformational preferences of glycine in the context of a folded protein, on the other hand, have received far less attention. The original Ramachandran plot for glycine, based on hardsphere steric repulsions, suggested that most of the (f,y) conformational space apart from the two axes (f= 0, y = 0) should be available to glycine, [9] but as the number of determined high-resolution protein structures has grown, a clear bimodal distribution of preferred (f,y) conformations, characterized by y = 08 AE 508 and y = 1808 AE 608, has emerged (Figure 1). Ho and Brasseur have recently analysed the discrepancy between theoretical and actual glycine conformational density and concluded that this bimodal distribution is a result of steric clashes between glycine a H i atoms and nearby O i and N i+1 atoms. [10] Our interest in glycine stems from a little-heralded observation about the frequency of glycine-proline (Gly-Pro) pairings at the C termini (CT) of helices. Gunasekaran et al. report that out of the 318 helices ending with a glycine in a left-handed helical (a L ) conformation in their dataset, none was found to have a proline immediately following the capping glycine, despite proline's predominance in this second capping position.[11] The authors attributed this curious finding to the inability of proline to form an amide hydrogen bond characteristic of a L -capped helices.[11] These findings appear to be a special case of a more widespread phenomenon reported two decades ago, in which no glycine residues were found to adopt the a L conformation when followed by a proline, regardless of secondary structural environment. [12] This phenomenon was attributed to a steric clash between Gly NH and Pro d CH 2 atoms. [12] In this work we re-examine the conformational preferences of glycine and the instances of ...