I think the PDB (and Gerard DVD) got the answer with their quite new validation report provided for each entry.On the other hand PDBredo (Tassos & Co) is an excellent tool to compare a currently deposited structure and its reprocessed version (although it doesn't tell about the phasing). Best,Philippe Philippe BENAS, Ph.D. Dog in the manger "Un importun survient qui trouble l'intimité, qui arrête l'expansion, qui glace le plaisir, - probablement comme un étranger tombant au milieu d'enfants en train de danser une ronde", Alfred Delvau, Dictionnaire de la langue verte (1866).
Laboratoire de Cristallographie et RMN Biologiques, UMR 8015 CNRS Faculté de Pharmacie, Université Paris Descartes Case 48 Av, de l'Observatoire F-75270 PARIS cedex 06 +33.1.5373.1599 E-mails: philippe.be...@parisdescartes.fr, philippe_be...@yahoo.fr URLs: http://lcrbw.pharmacie.univ-paris5.fr/ , http://lcrbw.pharmacie.univ-paris5.fr/spip.php?article18 De : "Goldman, Adrian" <adrian.gold...@helsinki.fi> À : CCP4BB@JISCMAIL.AC.UK Envoyé le : Mardi 27 juin 2017 10h59 Objet : Re: [ccp4bb] Incorrect Structure in the PDB I agree: I don’t think this is fraud, and was never even suggesting it. If it were fraud, _I_ at least would back-transform my structure so that there weren’t horrible red and green blobs all over the place…! I think it’s just sloppy - and the question remains, as posed by the original poster: what and how to do about the many examples? Adrian On 27 Jun 2017, at 11:55, Bernhard Rupp <hofkristall...@gmail.com> wrote: I beg to differ. You are not pulling a murthy simply by depositing a poor or even wrong structure. Although the borderline beteeen sloppy work (and selfdeception) and reckless misleading of others can be floating, real cases of fraud and fabrication are almost exceptionally rare. Best, br On Jun 27, 2017 10:44, "Philippe BENAS" <00000d88e888355a-dmarc-requ...@jiscmail.ac.uk> wrote: Dear Adrian, OK, I understand. You might be perfectly right. Moreover as I wrote it is difficult to tell something from a single mono view picture.And as you are pointing out their Ramachadran plots doe not look good at all. So they are poor crystallographers. But the question that remains is to know if their structure is really wron, I mean if the backbone is incorrect or not. For sure they might end up with a poor reputation as crystallographers if they publish so badly refined structures. But they might end up asKrishnaMurthy if the backbone of their published structures are wrong ! All the best,Philippe Philippe BENAS, Ph.D. Dog in the manger "Un importun survient qui trouble l'intimité, qui arrête l'expansion, qui glace le plaisir, - probablement comme un étranger tombant au milieu d'enfants en train de danser une ronde", Alfred Delvau,Dictionnaire de la langue verte (1866). Laboratoire de Cristallographie et RMN Biologiques, UMR 8015 CNRS Faculté de Pharmacie, Université Paris Descartes Case 48 Av, de l'Observatoire F-75270 PARIS cedex 06 +33.1.5373.1599 E-mails:philippe.benas@parisdescartes. fr, philippe_be...@yahoo.fr URLs: http://lcrbw.pharmacie.univ- paris5.fr/ , http://lcrbw.pharmacie.univ- paris5.fr/spip.php?article18 De : "Goldman, Adrian" <adrian.gold...@helsinki.fi> À : Philippe BENAS <philippe_be...@yahoo.fr> Envoyé le : Mardi 27 juin 2017 10h30 Objet : Re: [ccp4bb] Incorrect Structure in the PDB Philippe All I did was take their deposited mtz and coordinates. The peak in question in green is about 7 sigma. This structure (at 2 Å) has 3% ramanachandran-forbidden; one of their lower-resolution structures is 22%! And no, given that the action of a fourier transform hasn’t changed, this is not the best structure that could have been built at that time (2002…). Adrian On 27 Jun 2017, at 11:22, Philippe BENAS <philippe_be...@yahoo.fr> wrote: Dear Adrian, dear all, It does not look very clear to me that the side chain for this tyrosine is at a wrong place for instance (indeed Chi2 torsion angle could have multiple values = highly rotating aromatic ring). But it seems rather that the main chain is not correctly traced, which is probably worse. In addition it is difficult to tell something from just a mono view... This being said data collected by one group or another vary. For instance, resolution and hence R/free R can (generally) be improved. This does not imply that the structure coming from the lower resolution structure in wrong: it is just probably the best model that could have been built with the available data at that time.And on the contrary to other techniques we, crystallographers, have the chance of the Fourier transform which permits to have statistical indicators, such as R and free R, which give an indication about the accuracy of the reported structure. Similarly when solving the structure by experimental phasing we have also such indicators (e.g. the figure of merit, Rcullis, ...).All these statistical factors (including those related to the collected I(hkl) themselves) allow to give more or less confidence to a deposited model. Now there are cases where people report a modelled part in their structure and this sounds fair to me as long as it is correctly stated in the publication and the PDB file. Another scenario would be a poorly resolved, I mean with erroneous phases, and this is probably the most tricky case (too long to debate in this email) And finally the last case corresponds to real scientific frauds such as the C. Murthy's case and that in principle end with multiple retractions, PDB withdrawals and a broken scientific career (even if take 15 years to get the crook). And such structures must be reported to the PDB for sure ! Best regards,Philippe Philippe BENAS, Ph.D. Dog in the manger "Un importun survient qui trouble l'intimité, qui arrête l'expansion, qui glace le plaisir, - probablement comme un étranger tombant au milieu d'enfants en train de danser une ronde",Alfred Delvau,Dictionnaire de la langue verte (1866). Laboratoire de Cristallographie et RMN Biologiques, UMR 8015 CNRS Faculté de Pharmacie, Université Paris Descartes Case 48 Av, de l'Observatoire F-75270 PARIS cedex 06 +33.1.5373.1599 E-mails:philippe.benas@parisdescartes. fr, philippe_be...@yahoo.fr URLs: http://lcrbw.pharmacie.univ- paris5.fr/ , http://lcrbw.pharmacie.univ- paris5.fr/spip.php?article18 De : "Goldman, Adrian" <adrian.gold...@helsinki.fi> À :CCP4BB@JISCMAIL.AC.UK Envoyé le : Mardi 27 juin 2017 9h36 Objet : Re: [ccp4bb] Incorrect Structure in the PDB There are plenty of such structures in the PDB. We have one that we are rerefining at the moment - and indeed there is a whole slew of structures from the same author, all done sloppily. Here’s an example of a 2 Å (!) structure. The Tyr clearly goes into the green blob above it, and the loop conformation is all wrong. I think the issue may be that there are too many people out there “solving structures” who actually have very little skill at doing so. Adrian loopth <Screen Shot 2017-06-27 at 10.29.51.png> On 27 Jun 2017, at 10:15, Trevor Sewell <trevor.sew...@uct.ac.za> wrote: The misinterpretation is considerable I as can be seen from the attached coot screenshot. I have no reason to suspect malfeasance. But it looks like the authors didn’t check very carefully.I have re-interpreted and refined the density and it is just fine – Rfactor of 18% for a 2.3A structure. The critical reinterpretations concern the orientation of the backbone near the active site and the interpretation of a blob of density claimed to be substrate in the original paper. Maybe the best would be to write to the author and suggest that she obsolete the structure. We could see if we could reach some agreement on how to take it further – perhaps a letter to the editor of JBC. Sent from Mail for Windows 10 From: Manfred S. Weiss Sent: Tuesday, June 27, 2017 8:46 AM To: Trevor Sewell Subject: Re: [ccp4bb] Incorrect Structure in the PDB Dear Trevor, you can download the incorrect structure and the associated data and reinterpret and re-refine the structure. Then you can re-deposit provided you write a paper about the new findings. This is currently the policy of the PDB. Else, you can contact the authors of the incorrect structure and do the reinterpretation together with them? They can replace the incorrect structure without a new publication. That's all there is at the moment. May I ask what is incorrect about the structure? Cheers, Manfred Am 27.06.2017 um 08:34 schrieb Trevor Sewell: I have come across a key paper in my field that describes an enzyme mechanism. Their work is based on a deposited structure – by other authors - that is incorrectly interpreted. Is there a process for removing a demonstrably wrong structure (deposited by others) from the PDB and replacing it with a correctly interpreted structure based on the original data? Or is there an alternative, and generally recognized, way of getting the correct structure in the public domain? Many thanks for your advice on this matter. Trevor Sewell Disclaimer - University of Cape Town This e-mail is subject to UCT policies and e-mail disclaimer published on our website at http://www.uct.ac.za/about/ policies/emaildisclaimer/ or obtainable from +27 21 650 9111. If this e-mail is not related to the business of UCT, it is sent by the sender in an individual capacity. Please report security incidents or abuse via cs...@uct.ac.za -- Dr. Manfred S. Weiss Macromolecular Crystallography Helmholtz-Zentrum Berlin Albert-Einstein-Str. 15 D-12489 Berlin Germany Helmholtz-Zentrum Berlin für Materialien und Energie GmbH Mitglied der Hermann von Helmholtz-Gemeinschaft Deutscher Forschungszentren e.V. Aufsichtsrat: Vorsitzender Dr. Karl Eugen Huthmacher, stv. Vorsitzende Dr. Jutta Koch-Unterseher Geschäftsführung: Prof. Dr. Bernd Rech (kommissarisch), Thomas Frederking Sitz Berlin, AG Charlottenburg, 89 HRB 5583 Postadresse: Hahn-Meitner-Platz 1 D-14109 Berlin http://www.helmholtz-berlin.de Disclaimer - University of Cape Town This e-mail is subject to UCT policies and e-mail disclaimer published on our website at http://www.uct.ac.za/about/ policies/emaildisclaimer/ or obtainable from +27 21 650 9111. If this e-mail is not related to the business of UCT, it is sent by the sender in an individual capacity. Please report security incidents or abuse via cs...@uct.ac.za <Around residue 228.png> <Screen Shot 2017-06-27 at 10.29.51.png>
