Figure 2
(a) KDE of C—N bond lengths in helix (light green) and strand (dark red) conformations. The density distributions demonstrate a slight difference in the peak positions of C—N bond lengths, with helices exhibiting a slightly higher peak density at approximately 1.32–1.33 Å compared with strands. The significant overlap in distributions indicates comparable bond lengths between the two secondary structures. (b) The backbone geometry of the central residue (residue 0) is depicted, including atoms from the adjacent residues (−1 and +1) that form its two peptide units. Seven bond angles associated with residue 0 are labelled, with ideal values listed from four key references in the descending order Corey & Donohue (1950 ), Engh & Huber (1991 , 2001 ) and Berkholz et al. (2009 ). Modern refinement and modelling programs commonly use the 1991 or 2001 Engh and Huber values or slight variations thereof. Rotatable bonds defining the backbone torsion angles ω, φ and ψ are also shown. The figure, adapted from Berkholz et al. (2009 ), incorporates updates from the current work, with new data for helices highlighted in green and strands in red. Additionally, standard deviations for each bond angle in helices and strands have been included to provide a quantitative measure of structural variability in each secondary structure. (c) KDE plot of seven key bond angles associated with peptide geometry. The top row represents the distributions of bond angles in helices, while the bottom row corresponds to those in strands. Each plot is based on data from a nonredundant set of 1024 high-resolution protein structures. The distributions illustrate distinct conformational preferences, highlighting the characteristic differences between helices and strands. Each plot includes a vertical dashed line (axvline) indicating the mean value of the angle, providing a visual representation of central tendencies within helices and strands. These mean values are indicated in Fig. 2 (b). |