Validate cif file

A structure report can also be based on massaged or artificial data, as exemplified by data archived with the CSD. Those cases need not always to be fraudulent but rather related to poor chemical knowledge and inexperience with the available options in software packages.

Software packages may include options to create artificial reflection data meant to be used for test purposes only. The residual-density ranges, as reported in the name. Those values are expected to be closely identical within rounding errors. Three sets of R , wR 2 and S values are calculated and compared. The first set is again calculated from the name. The second set is based on the observed and calculated reflection data in the name.

The third set are the corresponding values as reported in the name. The ALERT implies that the H-atom positions are not necessarily to be trusted for detailed hydrogen-bond analysis. ALERT indicates that the embedded name. A non-zero value of ALERT can be taken as a measure of the quality of the difference density map of a refined structure assuming that spherical atomic scattering factors were used.

That is left to human beings, either referees or users of the reported results. The value of a structure report lies in the added reliable scientific information. A well-documented poor-quality structure may be fine if supporting part of the reported chemistry in a chemical journal, but less suitable for publication in a crystallographic journal such as Acta Crystallographica that aims to publish the best attainable quality of the reported structure, the latter in view of future use of the data in unrelated research such as non-AIM model refinement.

Not all reported issues with a structure report can be resolved with access to the reflection data. Some can be resolved only with access to the diffraction images. Work on the archiving of those data is underway Kroon-Batenburg et al. Refinements against powder diffraction data e. A special version of the software is available without that dependency.

The keyboard instruction platon -u name. The alternative command platon -U name. I am also indebted to many who supplied ideas, comments and in particular CIFs with new issues to be addressed. Betteridge, P. Acta Cryst. E 75 , — C 73 , — D 66 , — E 74 , — C 65 , o—o A 39 , — IUCrJ , 5 , 32— B 72 , — A 37 , — A 47 , — A 32 , — A 65 , — Report ORNL Google Scholar Kroon-Batenburg, L.

IUCrJ , 4 , 87— A 75 , 50— B 57 , — IUCrJ , 1 , 49— Crystal Structure Refinement. Oxford University Press. Google Scholar Palatinus, L. B 69 , — Google Scholar Sheldrick, G. A 64 , — A 71 , 3—8. C 71 , 3—8. A 46 , — D 65 , — C 71 , 9— Acta , , — This is an open-access article distributed under the terms of the Creative Commons Attribution CC-BY Licence , which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.

Issue contents. Download PDF of article. Navigation 1. Introduction 2. Experimental data and data reduction 3. Structure solution 4. Refinement 5. Validation 6. Concluding remarks 8. Article statistics. Next article. Volume 76 Part 1 January Pages Open access. Anthony L. Edited by H. Similar articles. Introduction Developments in diffractometer technology and software have made the collection of X-ray diffraction data on single crystals rapid and routine.

Experimental data and data reduction Most current crystal structure reports are based on data collected with area detectors as opposed to previously used serial detectors. Experimental data It is good practice to collect diffraction data up to at least copper sphere resolution, i. Data reduction 2. Absorption correction Given a crystal with faces that can be indexed and bathed fully in the homogeneous part of an X-ray beam, there are numerical and analytical ways to correct for absorption.

Symmetry The symmetry of the electron density of a crystal structure can, in the majority of reported cases, be associated with one of the distinct three-dimensional symmetry patterns. Structure solution The path from a preliminary structure to a refined structure is full of pitfalls for the unwary.

Connected set Preliminary structure solution does not always result in a logically ordered, sensibly labelled and connected set of atoms in the asymmetric unit. Missed symmetry The applicable symmetry of a crystal structure is not always obvious.

Twinning detection checkCIF includes a test for twinning. Refinement The experimental electron density in a crystal structure is approximated by a set of structure model parameters. Constraints and restraints Most of the model parameters can be refined.

Figure 1 Example of hydrogen atoms wrongly calculated in sp 2 rather than sp geometry positions. Wrongly calculated hydrogen-atom positions show up as negative red densities in the difference density map and correct places as positive green densities. Contours are drawn with 0. Figure 2 a Difference density map for a structure with wrong atom-type assignments. Figure 3 a Example of a structure with infinite channels filled with unresolved residual density and b display of the infinite channel, artificially filled with spheres.

Figure 4 a Truncation ripples in an F obs map and b an essentially clean associated difference-density map. Figure 5 Difference-density map with residual density maxima on bonds for a good structure refined with spherical atomic scattering factors AIM model.

Convergence quality criteria A low R value is frequently used as a criterion for a good structure. Hirshfeld test The Hirshfeld rigid-bond test Hirshfeld, was originally not introduced as a validation test but turned out to be also useful as a validation tool. Occupancy and multiplicity The occupancy factor as documented in the CIF is sometimes confused with the site-occupancy factor SOF as used in refinement programs such as SHELXL that include a weighting factor for atoms in special positions.

The unmerged reflection file An archived CIF is expected to include the unmerged reflection data on which a study is based. References Betteridge, P. Follow Acta Cryst. The IUCr is a scientific union serving the interests of crystallographers and other scientists employing crystallographic methods.

Search IUCr Journals. Positive or negative densities on non-hydrogen-atom sites may indicate wrong atom-type assignments Fig. Thoughtful examination of the model and possible consultation with synthetic chemists will help to resolve them. Example of hydrogen atoms wrongly calculated in sp 2 rather than sp geometry positions. Wrongly calculated hydrogen-atom positions show up as negative red densities in the difference density map and correct places as positive green densities.

Contours are drawn with 0. Other issues in the difference-density map may easily escape attention or be explained away as being due to poor data when one just relies on the maximum and minimum residual density excursions reported by the refinement software. Some programs do not report difference-map density on atom sites.

Contoured two- and three-dimensional difference-density maps offer a more detailed perspective. Significant approximately spherical density in a difference-density map cannot be ignored.

Disordered solvent sites, in particular when occupied by a mixture of solvents or solvents incommensurately ordered in infinite channels Fig. Algorithms, such as the one used in checkCIF , are available to detect those regions in the structure as solvent-accessible voids. These are often artifacts related to improper correction for absorption and similar systematic effects. They should not be confused with or explained as arising from positive and negative ripples around heavy atoms, related to resolution truncation, as might be seen in F obs maps Fig.

Absorption artifacts usually show up as a symmetrical pattern of positive and negative peaks around the heavy atoms, correlated with the external shape of the crystal. Structure reports for organic compounds based on good data and refined with the AIM model usually have their largest residual electron density on bonds Fig.

Poor data result in noisy difference maps. Difference-density map with residual density maxima on bonds for a good structure refined with spherical atomic scattering factors AIM model.

A low R value is frequently used as a criterion for a good structure. Unfortunately, that goal is often achieved in structure reports by limiting the resolution of the data used in the refinement or by removing outliers without proper justification.

Other parameters such as the data-to-parameter ratio, wR 2 and S values and weighting parameter values have to be considered as well. A refinement program such as SHELXL optimizes two parameter values in the reflection-weighting expression to achieve an S value close to 1.

Failure to come close to that value might be indicative of unresolved issues in the data or the model. In addition large values, in particular for the second parameter, in the SHELXL weighting expression may convey a similar message. A non-Gaussian error distribution of the observed and calculated intensities may be visualized with a normal probability plot included in the name.

The central part of the plot should be straight and plot not too S-shaped, indicating many outliers. Significant deviations of the scaling K and GooF values from 1. Details can be found in the name. A sufficient reflection data N to refined parameter p ratio is expected. The number of reflections in this ratio is the number of Laue group averaged reflections.

The checkCIF test criteria for the N : p ratio are set on values that can be met with good data for both centrosymmetric and non-centrosymmetric structures. As many Bijvoet pairs as possible will be needed for a reliable refinement of that parameter. Usually, as a first step to establish a proper value for the Flack parameter, the non-centrosymmetric structure is refined without this refineable parameter included.

Subsequently a post-refinement analysis is performed with the observed and calculated Bijvoet differences to estimate the value of the Flack parameter with associated estimated standard uncertainty.

Various estimates for that are in use such as the Parsons et al. Based on this analysis, subsequent refinement will be in order, including the Flack x parameter, when the estimated values deviate significantly from zero, including proper inversion of the coordinates when greater than 0. The name. A name. The result of that analysis is written to the name. By default, a built-in version of that file is used. However, it is even better to refer to some explanatory text in the report.

I, II, etc. C and E can be included preferably as the first block. In cases where some of the atoms are located on special positions or with polymeric structures, things are less well defined.

The main requirement is consistency with associated data. The notion is that the components of the displacement parameters of the two atoms in a covalent bond have approximately equal opposite values along that bond. This notion was extended to the generally weaker coordination bonds to a metal and transformed into a validation test.

In this test, the Hirshfeld difference divided by its associated s. Large deviations may point to erroneously mis-assigned atom types. This test turned out to be instrumental in discovering published structures with deliberately changed atom types.

Values for other wavelengths such as those for a synchrotron data collection will need to be user supplied. For example, a SOF value of 0. This might be due to a wrong hybridization assignment or putting the CH 3 moiety in the wrong staggered or eclipsed orientation or not carrying out a rigid-rotating group refinement for the methyl group.

Other not uncommon cases are the mis-assignment of a non-hydrogen atom carrying a hetero atom e. S or N and a C—H in a hetero ring moiety.

Exchanging the assignments should solve the issue. Short contacts involving disordered atom sites should be inspected in detail to ensure that they are physically possible and are not artefacts of the disorder model. Finally there are details listed for the check for twinning, absolute structure and a detailed analysis of the peaks in the final difference map.

An archived CIF is expected to include the unmerged reflection data on which a study is based. That file can be used to investigate unresolved issues concerning the correct space-group assignment. A merged reflection file has lost information on the quality and data consistency of systematic absences and indications for the lack of a centre of symmetry and absolute structure. Unfortunately, the unmerged reflection file is already several steps away from the primary experimental data: the diffraction images.

Data reduction integration, correction for absorption etc. Information about scattering effects outside the integrated diffraction spots and their shape is easily lost but often relevant to understand the origin of a poor structure refinement.

A structure report can also be based on massaged or artificial data, as exemplified by data archived with the CSD. Those cases need not always to be fraudulent but rather related to poor chemical knowledge and inexperience with the available options in software packages.

Software packages may include options to create artificial reflection data meant to be used for test purposes only.

The residual-density ranges, as reported in the name. Those values are expected to be closely identical within rounding errors. Three sets of R , wR 2 and S values are calculated and compared. The first set is again calculated from the name.

The second set is based on the observed and calculated reflection data in the name. The third set are the corresponding values as reported in the name. The ALERT implies that the H-atom positions are not necessarily to be trusted for detailed hydrogen-bond analysis.

ALERT indicates that the embedded name. A non-zero value of ALERT can be taken as a measure of the quality of the difference density map of a refined structure assuming that spherical atomic scattering factors were used. That is left to human beings, either referees or users of the reported results. The value of a structure report lies in the added reliable scientific information. A well-documented poor-quality structure may be fine if supporting part of the reported chemistry in a chemical journal, but less suitable for publication in a crystallographic journal such as Acta Crystallographica that aims to publish the best attainable quality of the reported structure, the latter in view of future use of the data in unrelated research such as non-AIM model refinement.

Not all reported issues with a structure report can be resolved with access to the reflection data. Some can be resolved only with access to the diffraction images. Work on the archiving of those data is underway Kroon-Batenburg et al. Refinements against powder diffraction data e. A special version of the software is available without that dependency. The keyboard instruction platon -u name. The alternative command platon -U name.

I am also indebted to many who supplied ideas, comments and in particular CIFs with new issues to be addressed. National Center for Biotechnology Information , U. Acta Crystallogr E Crystallogr Commun. Published online Jan 1. Anthony L. Author information Article notes Copyright and License information Disclaimer. Correspondence e-mail: ln. Received Dec 2; Accepted Dec 2.

Spek This is an open-access article distributed under the terms of the Creative Commons Attribution CC-BY Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited. This article has been cited by other articles in PMC. Introduction Developments in diffractometer technology and software have made the collection of X-ray diffraction data on single crystals rapid and routine.

Experimental data and data reduction Most current crystal structure reports are based on data collected with area detectors as opposed to previously used serial detectors.

Experimental data It is good practice to collect diffraction data up to at least copper sphere resolution, i. Data reduction Absorption correction Given a crystal with faces that can be indexed and bathed fully in the homogeneous part of an X-ray beam, there are numerical and analytical ways to correct for absorption. Link uProc with the Start node. AppyPie automates your business flow with any application you currently use in your workday.

Please, access to uProc integration to create workflows at AppyPie. Create another step and search for uProc integration you will need to add your uProc account to AppyPie. Choose your preferred language and method to process your data one value, multiple values, or streams — files, databases, and cloud applications -.

If you have a lot of rows to be processed you can check our discounts at Purchase section , completing the next data:. How many rows do you want to process? You can make uProc better. Share your thoughts with us to add new tools or improve any other aspect of the platform. What can I do with this tool? If you use this tool you can discover if a cif card number is valid.

Only for Spain If you use the tool you will verify your data. Who can use this tool? Sample: A What fields can I get? Using the tool from web wizard Web wizard allows processing any file download sample template , database or cloud application using existing data tools.

Add tool to view the estimated cost and process it Boosting data from Google Sheets Before processing any data, check the next steps to get add-on working in your sheet: One logged account is required. Multiple accounts cause permission errors in this add-on. Logout from all of them and use only one account with Google. Install uProc for Sheets add-on from Google Apps marketplace with one connected account. You have to be the owner of the sheet.