Dear Sam, As already pointed out by other CCP4bbers, I think trying to merge non-overlapping resolution shells cannot lead to good data. In addition it is very straight forward and fast to remove ice-rings during data processing with either XDS or imosflm, as also already pointed out. And your I_obs and Sig_I_obs will be much more accurate.
HTH, Philippe Philippe BENAS, Ph.D. ARN UPR 9002 CNRS IBMC Strasbourg 15, rue René Descartes F-67084 STRASBOURG cedex +33.3.8841.7109 E-mails: p.be...@ibmc-cnrs.unistra.fr, philippe_be...@yahoo.fr URLs: http://www-ibmc.u-strasbg.fr/ , http://www-ibmc.u-strasbg.fr/spip-arn/ -------------------------------------------- En date de : Mar 9.4.19, Sam Tang <samtys0...@gmail.com> a écrit : Objet: Re: [ccp4bb] [EXTERNAL] Re: [ccp4bb] High Rfree - ice ring À: CCP4BB@JISCMAIL.AC.UK Date: Mardi 9 avril 2019, 13h27 Hello again. I agree the get-around strategy we took is not a good practice at all. For our initial imosflm run we actually turned on 'exclude ice ring' button. The following was reported in the log: ICE RING SUMMARY: reso ice_ring mean_I mean_Sigma Estimated_I Ratio Zscore Completeness Ave_Completeness 3.88 yes 39933.04 3690.30 1398.21 28.56 10.44 0.48 nan 3.67 yes 44809.76 4257.56 778.04 57.59 10.34 0.58 nan 3.43 yes 7270.25 885.61 532.75 13.65 7.61 0.54 nan 2.66 yes 2070.19 488.66 156.09 13.26 3.92 0.46 nan A total of >2200 reflections were already omited. The range of poor R seems to correlate to the range 3.8-3.6A. I am thus also thinking there may be other issues (not visibly identified on images) other than ice rings. We actually first merged the two datasets (high and low resolutions) in pointless before presenting to aimless. We are trying over other different strategies to see if we can get a better tackle. Will report again soon. Sam On Tue, 9 Apr 2019 at 18:47, Johan Turkenburg <00002a539df422fe-dmarc-requ...@jiscmail.ac.uk> wrote: I agree with Harry that an ice ring should never require you to process the data in two separate runs, and hopefully this does not become a standard approach.......... How did you present those data to aimless so it could scale the two datasets that have no overlap at all? Johan On Tue, 9 Apr 2019 at 10:07, Harry Powell <0000193323b1e616-dmarc-requ...@jiscmail.ac.uk> wrote: Hi Sam Did you use the ice-ring exclusion option in iMosflm (a button that has an image like a snowflake)? It should exclude data in _narrow_ resolution rings (substantially less than 0.2Å!) around the ice rings, and can be set for any combination of indexing, refinement and integration. There should not be any need to process the data twice, once for the low resolution data and once for the high. Harry -- Dr Harry Powell To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1 On 8 Apr 2019, at 19:50, Sam Tang wrote: Hello everyone Thanks a lot for your input and advices. To report on how we tackled the issue - (1) We used imosflm to integrate the data. (2) We eventually integrated the data in two resolution ranges, say 45A-3.5A, and 3.3A-3A, and merge them by Aimless. I must add that indeed from the log file for our initial round the program had already identified some ice ring regions. Aimless statistics looked fine and we were able to get a MR solution which was refined to much better Rf/Rw. This is definitely not a smart solution because we effectively 'throw away' useful data between 3.5A-3.3A, but for the purpose of MR and refinement, it seems we have solved (or simply bypassed?) the problem. Suggestions on XDS/DIALS are appreciated. We are actually using this dataset as a test set for XDS/DIALS to deal with ice rings. Will further report if we've got anything interesting. Thanks again! Sam On Thu, 4 Apr 2019 at 20:54, Clemens Vonrhein <vonrh...@globalphasing.com> wrote: Dear all, And if you want to process with XDS: autoPROC [1] will try to detect and exclude ice-rings automatically - if present [2]. If you know that you have ice-rings you can force it [3] to exclude all known ice-rings ranges - but this might not be the best solution if you have "just" diffuse ice-rings (where the special treatment of background within DIALS might be better). Something to test and compare maybe? Cheers Clemens [1] https://www.globalphasing.com/autoproc/ https://www.globalphasing.com/autoproc/wiki/index.cgi?IceRingHandling [2] https://www.globalphasing.com/autoproc/manual/autoPROC7.html#step1_spotnohkl https://strucbio.biologie.uni-konstanz.de/xdswiki/index.php/Ice_rings [3] https://www.globalphasing.com/autoproc/manual/appendix1.html#SetvarParameter_XdsExcludeIceRingsAutomatically On Thu, Apr 04, 2019 at 10:51:19AM +0000, melanie.voll...@diamond.ac.uk wrote: > Dear Sam, > > > to continue from James Parkhurst's email... > > > You can do more analysis regarding ice rings using Auspex (https://www.auspex.de/) if you already have some integrated file. > > Regarding re-integrating images, what did you use the first time round? I think if you use DIALS it got some clever implementation in it that is better in estimating the errors of reflections in proximity to ice rings. Hence you should get better Rfactors without having to remove the effected resolution range and the data it covers (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5619854/). > > > HTH > > > M > > > ________________________________ > From: CCP4 bulletin board <CCP4BB@JISCMAIL.AC.UK> on behalf of herman.schreu...@sanofi.com <herman.schreu...@sanofi.com> > Sent: 04 April 2019 10:26:09 > To: ccp4bb > Subject: [ccp4bb] AW: [EXTERNAL] Re: [ccp4bb] High Rfree - ice ring > > > Dear Sam, > > > > I would remove the ice ring and reprocess the data. Ice rings may wreak havoc with scaling so at minimum you have to redo the scaling. > > > > Best, > > Herman > > > > Von: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] Im Auftrag von Sam Tang > Gesendet: Donnerstag, 4. April 2019 11:01 > An: CCP4BB@JISCMAIL.AC.UK > Betreff: [EXTERNAL] Re: [ccp4bb] High Rfree - ice ring > > > > > Dear Eleanor and Eric > > > > Thanks for your replies. > > > > Yes indeed when we looked at the plots e.g. R factor vs resln there was a sharp peak near 3.6 - 3.8 A which is where we visibly saw an ice ring on the image. Thus our first thought was to remove the ic ring. (either reprocess or can we bypass this resolution range during refinement?) > > > > The protein is 50 kDa, two molecule in the ASU, seemingly no obvious density was unassigned. We got ~30000 total observations, ~15000 unique observations. NCS restraints was applied. > > > > Best regards > > Sam > > > > > > > > On Thu, 4 Apr 2019 at 08:57, Eric Montemayor > <montemayor.e...@gmail.com<mailto:montemayor.e...@gmail.com>> wrote: > > That’s a rather large gap between Rwork and Rfree. I suspect you have mis-assigned your space group and as a result have a large number of copies in your asymmetric unit. Any structure can be solved in P1, but that does not mean the true space group is indeed P1. If you use P1 when it’s not actually P1, you will have an unnecessarily overparamerized model, hence the large gap between Rwork and Rfree. > > > > Questions: > > 1- how many copies in your asymmetric unit in P1? > > 2- how many atoms in your model vs number of unique reflections? > > 3- if more than one copy per asymmetric unit, are you imposing NCS restraints during refinement? > > > > -Eric > > > > > > > > On Wed, Apr 3, 2019 at 1:41 PM Sam Tang > <samtys0...@gmail.com<mailto:samtys0...@gmail.com>> wrote: > > Hi everyone again > > > > Hmmm I think we have solved a structure in P1 space, to 2.5 A. However after refinement the Rfree stuck at 33%-35% with Rwork around 26%. The structure was solved by MR and current model seems to fit density well. In Refmac log I found that at the resolution corresponding to high R there may be a solvent/ice ring. Since imosflm should be able to exclude ice rings, I am not 100% sure whether it's the cause to high R. But if this is actually the case, is there a way I can exclude certain resolution bins during Refmac (and is it an appropriate way to do so?) > > > > PS - the data is not affected by twining or pseudosymmetry as checked by Xtriage. > > > > Many thanks! > > > > Sam > > > > ________________________________ > > To unsubscribe from the CCP4BB list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1<https://urldefense.proofpoint.com/v2/url?u=https-3A__www.jiscmail.ac.uk_cgi-2Dbin_webadmin-3FSUBED1-3DCCP4BB-26A-3D1&d=DwMFaQ&c=Dbf9zoswcQ-CRvvI7VX5j3HvibIuT3ZiarcKl5qtMPo&r=HK-CY_tL8CLLA93vdywyu3qI70R4H8oHzZyRHMQu1AQ&m=UECB-BTblxRTCrEA2arfPs5VNLr4ntyGxszsw_5Jaxc&s=QloAn3zc13ZQX64uxy6UZygomHXUigCIkpTw8ZhYpbk&e=> > > > > ________________________________ > > To unsubscribe from the CCP4BB list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1<https://urldefense.proofpoint.com/v2/url?u=https-3A__www.jiscmail.ac.uk_cgi-2Dbin_webadmin-3FSUBED1-3DCCP4BB-26A-3D1&d=DwMFaQ&c=Dbf9zoswcQ-CRvvI7VX5j3HvibIuT3ZiarcKl5qtMPo&r=HK-CY_tL8CLLA93vdywyu3qI70R4H8oHzZyRHMQu1AQ&m=UECB-BTblxRTCrEA2arfPs5VNLr4ntyGxszsw_5Jaxc&s=QloAn3zc13ZQX64uxy6UZygomHXUigCIkpTw8ZhYpbk&e=> > > ________________________________ > > To unsubscribe from the CCP4BB list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1 > > -- > This e-mail and any attachments may contain confidential, copyright and or privileged material, and are for the use of the intended addressee only. If you are not the intended addressee or an authorised recipient of the addressee please notify us of receipt by returning the e-mail and do not use, copy, retain, distribute or disclose the information in or attached to the e-mail. > Any opinions expressed within this e-mail are those of the individual and not necessarily of Diamond Light Source Ltd. > Diamond Light Source Ltd. cannot guarantee that this e-mail or any attachments are free from viruses and we cannot accept liability for any damage which you may sustain as a result of software viruses which may be transmitted in or with the message. > Diamond Light Source Limited (company no. 4375679). Registered in England and Wales with its registered office at Diamond House, Harwell Science and Innovation Campus, Didcot, Oxfordshire, OX11 0DE, United Kingdom > > > ######################################################################## > > To unsubscribe from the CCP4BB list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1 -- *-------------------------------------------------------------- * Clemens Vonrhein, Ph.D. vonrhein AT GlobalPhasing DOT com * Global Phasing Ltd., Sheraton House, Castle Park * Cambridge CB3 0AX, UK www.globalphasing.com *-------------------------------------------------------------- ######################################################################## To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1 To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1 To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1 -- Dr. Johan P. Turkenburg X-ray facilities manager York Structural Biology Laboratory University of York York YO10 5DD UK Phone (+) 44 1904 328251http://orcid.org/0000-0001-6992-6838EMAIL DISCLAIMER http://www.york.ac.uk/docs/disclaimer/email.htm To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1 To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1 ######################################################################## To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1
