A distance geometry-based description and validation of protein main-chain conformation

Abstract

Understanding the protein main-chain conformational space forms the basis for the modelling of protein structures and for the validation of models derived from structural biology techniques. Presented here is a novel idea for a threedimensional distance geometry-based metric to account for the fine details of protein backbone conformations. The metrics are computed for dipeptide units, defined as blocks of C iÀ1 -O iÀ1 -C i -O i -C i+1 atoms, by obtaining the eigenvalues of their Euclidean distance matrices. These were computed for $1.3 million dipeptide units collected from nonredundant good-quality structures in the Protein Data Bank and subjected to principal component analysis. The resulting new Euclidean orthogonal three-dimensional space (DipSpace) allows a probabilistic description of protein backbone geometry. The three axes of the DipSpace describe the local extension of the dipeptide unit structure, its twist and its bend. By using a higher-dimensional metric, the method is efficient for the identification of C atoms in an unlikely or unusual geometrical environment, and its use for both local and overall validation of protein models is demonstrated. It is also shown, for the example of trypsin proteases, that the detection of unusual conformations that are conserved among the structures of this protein family may indicate geometrically strained residues of potentially functional importance.

High‐accuracy protein structure prediction in CASP14

“…20 The DipScore is a distance geometry-based metric and a local protein backbone validation score that is used for guiding automated protein model building with ARP/wARP at medium-to-low resolution 20,21 and as a general protein backbone validation score, 22 being also useful for the identification of geometrically strained residues of potentially functional importance. 20 Briefly, it is computed based on the all-to-all interatomic distances between the backbone Cɑ and O atoms of the residue to be analyzed and its two flanking neighbors, and evaluates the likelihood of the observed combination of interatomic distances to be correct based on what is found in high-quality, high-resolution structures from the PDB and those expected from a random sampling of atoms around a target Cɑ; the closer to one the DipScore is, the more likely that conformation is to be geometrically correct. It is computed using DipCheck, 20 distributed through the ARP/wARP and CCP4 23 software packages.…”

Elucidation of the viral disassembly switch of tobacco mosaic virus

“…Mg 2+ was placed in proximity of the RNA and justified by the known tendency of RNA to be stabilized by Mg 2+ ions. Validation scores were calculated with phenix.molprobity [39], phenix.em_ringer [40], and DipCheck [41]. The Ca 2+ ion was only found in the cryo-EM structure with high Ca 2+ concentrations.…”

“…Grouped atomic displacement factors (ADP) were refined with phenix.real_space_refine. Validation scores were calculated with phenix.molprobity [39], phenix.em_ringer [40], and DipCheck [41]. To assess overfitting of the refinement, we introduced random coordinate shifts into the final models using the program phenix.pdbtools with the shake option and a mean error of 0.5 Å , followed by refinement against the first unfiltered half-map (half-map 1) with the same parameters as above.…”

ALEPH: a network-oriented approach for the generation of fragment-based libraries and for structure interpretation

“…Distortions in the polypeptide chain are sometimes encountered, and especially when the resolution falls below 3-3.5 Å (Headd et al, 2012;Karmali et al, 2009) some structures may fail to meet DSSP regularity. DipSpace (Pereira & Lamzin, 2017) embeds geometrical information about the backbone atoms around each C atom in its dipeptide-unit environment, which is described as a matrix of the interatomic distances. Also, CaBLAM (Richardson et al, 2018) defines a novel parameter space of C -C and CO-CO virtual dihedrals, where the CO dimension diagnoses large distortions of peptide orientation at low resolution and the two C dimensions identify the probable secondary structure obscured by these problems.…”