Hi Pavel, Similar results in PDB-REDO https://pdb-redo.eu/db/5xql, nothing wrong but clear room for improvement. That said, I would hope that the epositors set the record straight on the data completeness or conjur up the missing reflection data.
Cheers, Robbie Sent from my Windows 10 phone Van: Pavel Afonine<mailto:[email protected]> Verzonden: donderdag 29 juni 2017 20:23 Aan: [email protected]<mailto:[email protected]> Onderwerp: Re: [ccp4bb] Questionable Data collection and refinement statistics for 5XQL I agree with Dominika, I can't see major problems with this entry (I also did some quick refinement and briefly looked at maps). Reported R factors match re-calculated values using data from PDB, which is good. Smaller issues I see are: - no solvent (water) in the model, while density suggests some; - maps don't look like 2.5A resolution, which is perfectly expectable given data completeness (overall 81%): Resolution Compl 35.051-13.721 89.38 13.660-11.061 95.19 11.024-8.858 98.41 8.848-7.119 100.00 7.115-5.723 100.00 5.714-4.596 100.00 4.595-3.692 100.00 3.692-2.967 95.92 2.966-2.494 56.98 and effective resolution along axes a,b,c: 2.453 2.453 3.167 A. - model contains 4% of rotamer outliers, some of them can be trivially fixed. - Xtriage shows three yellow "traffic lights". I'm sure many PDB entries (especially older ones) would have something that you could do better or fix using modern tools. Pavel On Thu, Jun 29, 2017 at 10:48 AM, Dominika Borek <[email protected]<mailto:[email protected]>> wrote: Dear C-Daddy, Did you look at the maps? Do you see any astonishing misinterpretations there? Do you think that the completeness in the last resolution shell would affect the maps and the model? If so, why? There are many possible reasons for the discrepancy you pointed out. In I422 a=b, alpha=90.00, so typos in the table are the most probable reason. I did look at the maps after running a quick refinement. Both the structure and the ligand are fine. I would probably add solvent molecules myself, but some conservative crystallographers prefer not to do it, even at 2.5 A. In other words, there is nothing wrong with this deposit. I would even say that the level of sloppiness is below the average for this one. D. On 2017-06-29 11:07 AM, CDaddy wrote: Dear colleagues, I am really astonished by a recently released structure 5XQL. This structure was just published in BBRC with a super fast process (accepted overnight). Significant inconsistences were observed when I compared the Data collection and refinement statistics in the paper and at PDB website. The paper shows the crystal belong to I 4 2 2 with a=135.77, b=136.11, c=95.13 (Å) ; α=90.01. The resolution is 2.5Å with completeness of 99.7 (99.1). However when I checked the released structure I got a= b=135.94, c=95.13 (Å) ; α=90.0. The resolution is 2.5Å with completeness of 81.5 (44.0). The completeness of the highest resolution shell is so low that 2.5Å seems unreasonable. At first I thought that the authors could be rookies in structural biology. But the molecular replacement in this paper could be very difficult due to very low sequence identity between the search model and the final structure. It was unlikely that two new hands could solve it easily. Is there anyone who has this kind of experience and know why? With best wishes, Richard -- Dominika Borek, Ph.D. *** UT Southwestern Medical Center 5323 Harry Hines Blvd. *** Dallas, TX 75390-8816 214-645-9577<tel:214-645-9577> (phone) *** 214-645-6353<tel:214-645-6353> (fax)
