Fig. 5: Increased structural frustration in backbone ensembles generated from original design blueprints.
From: Role of backbone strain in de novo design of complex α/β protein structures
a Backbone structure ensembles generated from blueprints for R2×3_BP1 and Pl2×3_BP. Individual members of the ensembles are represented by wireframe and averages over the thousands of backbones generated for each blueprint, in tubes. b The β-sheet formation probability and packability between the N- and C- terminal helices for the blueprints used in the designs and the strand-swapped NMR-structure-based blueprints. c Cumulative probability distributions of the distance between the N- and C- terminal helices for Pl2×3_BP, R2×3_BP1_A5_NMR, R2×3_BP1_B9_NMR, and R2×3_BP1. Filled boxes represent the distances of design models and open boxes, those of NMR structures. d Structures generated from R2×3_BP1 with the distance between the N- and C- terminal helices 10, 11, and 12 Å demonstrate that close helix packing is achieved at less than ~11 Å (for visualization, valine side chains are placed on the backbones at all residue positions). e (top) Schematic diagrams of hydrogen bonds in the parallel β-sheets of the designs and NMR structures. (bottom) Hydrogen-bond (HB) formation probability for each hydrogen bond (The numbering in the bar graph corresponds to that shown in the β-sheet schematics). Red arrows show the incomplete hydrogen bonds, in which donor is provided by the first residue of the strands. The dotted hydrogen bond in R3×3_BP1_9_NMR was not included because this bond is broken both in the NMR structures and the simulation ensemble.
