Hi Satvik One thing I don't understand is why there are no dips in your plot for an ice ring around 3.6Å - for hexagonal ice it should be about as strong as the ring at 2.25Å. It would be very useful to actually see the images to try to work out what the real issue is.
I think it would be best to continue this discussion off the ccp4bb and to let someone with extensive data processing experience have a look at your images. This will probably be faster and more definitive in the long run. I will be meeting developers of both XDS and DIALS next week and we should have some time to spend on your problem if we have access to the original images! Prof Dodson will also be present so we will be able to make use of her experience. The bug concerning the specific resolution range exclusion appears to be quite long-standing, so I'm not surprised that it is in other versions of iMosflm. Harry -- Dr Harry Powell Chairman of International Union of Crystallography Commission on Crystallographic Computing Chairman of European Crystallographic Association SIG9 (Crystallographic Computing) On 10 Aug 2017, at 11:13, Satvik Kumar wrote: > Dear Prof. Powell, > > In order to get past the bug, I reprocessed the data using a different > installation of iMOSFLM. The option "automatic ice and powder ring exclusion" > was toggled ON and also the snowflake button was kept ON during indexing, > cell refinement and integration. I observe that these changes have improved > the statics at the cost of completeness. The mean I/sigI has increased from > -0.2 to 0.4 but the completeness has reduced from 98.2 to 85.5 (64.5 to 31.9, > outershell). > > I inspected the wilson plots carefully but I am unable identify if the > re-processing has helped (plots are attached). > > I am unable to manually exclude specific resolution shells even with this > different installation of iMOSFLM. > > Please share your feedback. > > Thanks once again. > > > Best, > Satvik > > > On Wed, Aug 9, 2017 at 6:15 PM, Harry Powell <hrp-ccp...@virginmedia.com> > wrote: > Hi >> I had processed the images using iMOSFLM. The option of “automatic ice and >> powder ring exclusion” was toggled ON when I processed the data. It is only >> now I realize that this is not the way to get rid of ice rings. >> > > This is due to the exclusion limits being set too conservatively for the ice > rings; you might consider it a bug, because this should be the way to get rid > of the ice rings! >> The latest paper on the use of iMOSFLM (Powell. H. R et al, Nature >> Protocols, 2017) suggests excluding data within specific resolution shells >> to get rid of the ice ring problem. I observe that if I set the limits >> 3.62-3.68, 2.23-2.26, 1.90-1.93 Å in “excluded resolution ranges” option of >> iMOSFLM, only the spots upto 3.6 Å are found and also predicted. Moreover >> all high resolution data is lost. Somehow I am not able to get this >> strategy working in iMOSFLM. >> > > This is due to a bug in the iMosflm code; it will be fixed in the next > release (I've told the current developer about it...). > > I could send you a fix so that this option works if you like. > > Harry > -- > Dr Harry Powell > Chairman of International Union of Crystallography Commission on > Crystallographic Computing > Chairman of European Crystallographic Association SIG9 (Crystallographic > Computing) > > > > On 9 Aug 2017, at 13:17, Satvik Kumar wrote: > >> Dear All, >> >> >> >> Thank you all for your inputs. >> >> >> >> You are all correct. The diffraction images have ice rings at 3.67, 2.24 and >> 1.9 Å. The intensity of these ice rings decrease with increasing resolution. >> In the Wilson plot, I clearly observe the spikes in intensity corresponding >> to these resolutions. >> >> >> >> >> >> The latest paper on the use of iMOSFLM (Powell. H. R et al, Nature >> Protocols, 2017) suggests excluding data within specific resolution shells >> to get rid of the ice ring problem. I observe that if I set the limits >> 3.62-3.68, 2.23-2.26, 1.90-1.93 Å in “excluded resolution ranges” option of >> iMOSFLM, only the spots upto 3.6 Å are found and also predicted. Moreover >> all high resolution data is lost. Somehow I am not able to get this >> strategy working in iMOSFLM. >> >> >> >> The other suggestion was to deice using AUSPEX or DEICE. The information >> available on the internet suggests AUSPEX is a diagnostic tool. Is it >> possible to use it to deice? I will be trying to get DEICE working shortly. >> >> >> >> Please share your thoughts as to how I should proceed. >> >> >> Thanks, >> >> Satvik >> >> >> On Tue, Aug 8, 2017 at 11:47 PM, Eleanor Dodson <eleanor.dod...@york.ac.uk> >> wrote: >> You have some horrible ice rings - some data processing software may be able >> to cut them out.. how are you processing it? >> Eleanor >> >> On 8 August 2017 at 15:43, Christian Roth <christianroth...@gmail.com> wrote: >> Your plots look strangely different to the old Scala output you posted >> before, but never mind. >> >> Paul is right that a negative intensity is not desired and your crystal has >> some issues with ice. >> >> That one icering around 2.26 must be massive taken into account how haywire >> your curve goes there. >> >> Have you had a look at the images? There should be something visible in that >> area. >> >> Christian >> >> >> Am 08.08.2017 um 15:17 schrieb Paul Emsley: >>> On 08/08/2017 15:07, Satvik Kumar wrote: >>>> Dear Prof. Powell and Prof. Dodson, >>>> >>>> Thanks for your reply and advise. >>>> >>>> As per your suggestion, I have re-scaled the intensities using Aimless at >>>> 1.861 A. >>>> >>>> I observe that the I/sigI has dropped to -0.8 >>> >>> That's not good. >>> >>> > and the behaviour of CC_1/2 is still anomalous. >>> >>> That made me laugh out loud. Perhaps not the best choice of adjective. >>> >>>> >>>> Also, when I inspect the Wilson plot (Fig. 1), I observe that the curve >>>> does not fall smoothly with respect to the reference curve (blue). Even >>>> with respect to one more Wilson plot from CCP4 website (Fig. 2), the curve >>>> from my aimless output is different and discontinuous. >>> >>> Icy! >>> >>> /me wonders if CCP4 are distributing auspex yet... >>> >>>> >>>> The second moment of I is constant only up to a resolution of 2.4 Å at a >>>> value of 3 (Fig. 3). I was not able to get some other plot to compare >>>> against mine. >>>> >>>> Please tell me if I can still go ahead and refine at 1.861 A. >>> >>> No you can't. >>> >>> Maybe with some chopping you can rescue some reflections beyond 2.1. >>> >>> Paul >>> >> >> >> >> <090817.pdf> > > > <100817_stats.pdf>