Carbohydrate 3D structure validation

Joosten, Robbie P.; Lütteke, Thomas

doi:10.1016/j.sbi.2016.10.010

Cited by 27 publications

(24 citation statements)

References 95 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, the major improvement observed over the last 15 years is probably due to the introduction of automated tools that are checking the accuracy of structural models deposited to the PDB 41,42 . However, the 1994‐1998 level of accuracy of β‐D‐GlcNAc units had still not been reached again in 2017‐2019, suggesting that the use of carbohydrate validation tools such as PDB‐CARE 43 or Privateer 24 may still need to be systematized 6 . Note, however, that neither PDB‐CARE nor Privateer presently check the cis ‐ trans state of GlcNAc units.…”

Section: Resultsmentioning

confidence: 99%

“…The growing availability of 3D protein‐carbohydrate complex structures allows to perform statistical studies of these interactions at the atomic level 4,5 . Unfortunately, the protein databank (PDB) contains numerous wrong carbohydrate annotations, 6 as well as severe structural defects, 7,8 due for instance to the lack of carbohydrate torsion control during crystallographic structure refinement 9 …”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Numerous severely twisted N‐acetylglucosamine conformations found in the protein databank

2020

View full text Add to dashboard Cite

Taking advantage of the known planarity of the N‐acetyl group of N‐acetylglucosamine, an analysis of the quality of carbohydrate structures found in the protein databank was performed. Few obvious defects of the local geometry of the carbonyl group were observed. However, the N‐acetyl group was often found in the less favorable cis conformation (12% of the cases). It was also found severely twisted in numerous instances, especially in structures with a resolution poorer than 1.9 Å determined between 2000 and 2015. Though the automated PDB‐REDO procedure has proved able to improve nearly 85% of the structural models deposited to the PDB, and does prove able to cure most severely twisted conformations of the N‐acetyl group, it fails to correct its high rate of cis conformations. More generally, for structures with a resolution poorer than 1.6 Å, it produces N‐acetylglucosamine models in slightly poorer agreement with experimental data, as measured using real‐space correlation coefficients. Significant improvements are thus still needed, at least as far as this carbohydrate structure is concerned.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Numerous severely twisted N‐acetylglucosamine conformations found in the protein databank

2020

View full text Add to dashboard Cite

show abstract

“…Importantly, this is not only a new resource in its own right, but it can also serve for better homology restraint generation for future structure refinement. It should be noted that the improvement over the entire resolution range is also the result of other improvements to the PDB‐REDO pipeline and the external programs therein, including better treatment of twinning, general improvements to TLS, NCS, and ADP refinement, validation, and correction of structural zinc sites, better handling of carbohydrates, improved selection of resolution cut‐off and the generation of anomalous difference maps when possible. All these developments are consistently and uniformly applied in all entries, in addition to the applicable homology‐derived restraints.…”

Section: Resultsmentioning

confidence: 99%

Homology‐based hydrogen bond information improves crystallographic structures in the PDB

et al. 2017

View full text Add to dashboard Cite

The Protein Data Bank (PDB) is the global archive for structural information on macromolecules, and a popular resource for researchers, teachers, and students, amassing more than one million unique users each year. Crystallographic structure models in the PDB (more than 100,000 entries) are optimized against the crystal diffraction data and geometrical restraints. This process of crystallographic refinement typically ignored hydrogen bond (H‐bond) distances as a source of information. However, H‐bond restraints can improve structures at low resolution where diffraction data are limited. To improve low‐resolution structure refinement, we present methods for deriving H‐bond information either globally from well‐refined high‐resolution structures from the PDB‐REDO databank, or specifically from on‐the‐fly constructed sets of homologous high‐resolution structures. Refinement incorporating HOmology DErived Restraints (HODER), improves geometrical quality and the fit to the diffraction data for many low‐resolution structures. To make these improvements readily available to the general public, we applied our new algorithms to all crystallographic structures in the PDB: using massively parallel computing, we constructed a new instance of the PDB‐REDO databank (https://pdb-redo.eu). This resource is useful for researchers to gain insight on individual structures, on specific protein families (as we demonstrate with examples), and on general features of protein structure using data mining approaches on a uniformly treated dataset.

show abstract

“…Importantly, it is not only a new resource in its own right, but it can also serve for better homology restraint generation for future structure refinement. In addition, the entire PDB-REDO databank is constructed with a single software version (which has not been possible before) and with new algorithms: these include better treatment of twinning, general improvements to TLS, NCS, and ADP refinement 1 , validation and correction of structural zinc sites 30 , better handling of carbohydrates 31 , improved selection of resolution cut-off and the generation of anomalous difference maps when possible. All these developments are consistently and uniformly applied in all entries, in addition to the applicable homology-derived restraints.…”

Section: Redo Pipeline (See Online Methods For Details)mentioning

confidence: 99%

Homology-based hydrogen bond information improves crystallographic structures in the PDB

Beusekom

Touw

Tatineni

et al. 2017

Preprint

View full text Add to dashboard Cite

Crystallographic structure models in the Protein Data Bank (PDB) are optimized against the crystal diffraction data and geometrical restraints. This process of crystallographic refinement typically ignored hydrogen bond (H-bond) distances as a source of information. However, H-bond restraints can improve structures, especially at low resolution where diffraction data are limited.To improve low-resolution structure refinement, we present methods for deriving H-bond information either globally from well-refined high-resolution structures from the PDB-REDO databank, or specifically from on-the-fly constructed sets of homologous high-resolution structures. Refinement incorporating HOmology DErived Restraints (HODER), improves geometrical quality and the fit to the diffraction data for many low-resolution structures. Using approximately 60 years of CPU-time in massively parallel computing, we constructed a new instance of the PDB-REDO databank, a novel resource to help biologists gain insight on protein families or on specific structures, as we demonstrate with examples.All rights reserved. No reuse allowed without permission.(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint . http://dx.doi.org/10.1101/147231 doi: bioRxiv preprint first posted online Jun. 7, 2017; 3 Crystallographic structure models are optimized against the crystallographic diffraction data and a priori known geometrical observations, the geometrical restraints. We are developing the PDB-REDO procedure, which among many decisions 1 optimizes the weight between crystallographic and geometrical observations 2 to re-refine and re-build macromolecular structures before 3 or after they are submitted to the PDB 4 . In PDB-REDO and any crystallographic refinement procedure however, low resolution diffraction data means that fewer observations of diffracted X-rays are available, and as resolution declines the crystallographic refinement problem becomes increasingly underdetermined 5. Restraint dictionaries 6,7 describing 'ideal' refinement targets for bond lengths, angles, planar groups and other well-defined chemical features, at low resolution become gradually insufficient to yield high-quality structure models. Additional, external restraints 8 can be defined and, for example, hydrogen bond restraints 9,10 (H-bonds) andRamachandran torsion angle restraints 9,11 have been used to enhance protein secondary structure quality, particularly at lower resolution.Macromolecular crystals diffract X-rays to higher or lower resolution in an unpredictable manner:even very similar proteins or the same protein bound to different ligands (e.g. drug candidates), can yield crystallographic data at different resolutions. This allows refinement methods to harvest information from a high-resolution "reference" model and use it to refine low-resolution models 9,10,[12][13][14][15] . Available implementations of this principle focus on harvesting restrai...

show abstract

Carbohydrate 3D structure validation

Cited by 27 publications

References 95 publications

Numerous severely twisted N‐acetylglucosamine conformations found in the protein databank

Numerous severely twisted N‐acetylglucosamine conformations found in the protein databank

Homology‐based hydrogen bond information improves crystallographic structures in the PDB

Homology-based hydrogen bond information improves crystallographic structures in the PDB

Contact Info

Product

Resources

About