 |
 |
Dear Professor Read and colleagues Thanks for your kind attention to the question. On closer look into the log file and the parameters, it happened to be that our project student mis-set IMEAN and SIGIMEAN to F and SIGF which mis-led the programme in tNCS correction. The run seems normal when we rectify this silly mistake (on which we've spent two whole afternoons trying to look out why!) Sorry for all the inconvenience. Kind regards Sam PhD candidate Biochemistry Programme, School of Life Sciences, CUHK On 20 April 2016 at 16:33, Randy Read <[log in to unmask]> wrote: > Dear Sam, > > As far as I know, changing the sequence identity or the number of copies > of the model shouldn't have any effect on the moments calculations in > Phaser. Changing the number of copies could change other things > downstream. For instance, if translational NCS (tNCS) is detected but the > number of copies you are searching for is not divisible by the number of > copies related by tNCS, the tNCS correction is not applied in the > calculation. Nonetheless, even then, the tNCS analysis is used in the > moments calculations. > > Could you send me the two complete log files (preferably offline) so I can > see what has happened here? > > Thanks! > > Best wishes, > > Randy Read > > > On 19 Apr 2016, at 12:32, Sam Tang <[log in to unmask]> wrote: > > > > Hello, > > > > I am carrying out molecular replacement on one of our recent data set > using Phaser-MR. The data set was indexed P1 to 2.5 Angstorm. In my initial > run, Phaser-MR returned the following: > > > > -------- > > TWINNING > > -------- > > > > tNCS/Twin Detection Table > > ------------------------- > > -Second Moments- --P-values-- > > Centric Acentric untwinned twin > frac <5% > > Theoretical for untwinned 3.00 2.00 > > Theoretical for perfect twin 2.00 1.50 > > Initial (data as input) 0.00 2.41+/-0.070 1 1 > > After Anisotropy Correction 0.00 2.12+/-0.052 1 1 > > After Anisotropy and tNCS ---n/a--- > > > > P-value < 0.01 for <5% twinned is considered significant > > Resolution for Twin Analysis (85% I/SIGI > 3): 3.54A (HiRes= 2.67A) > > > > > > Afterward the initial run I tried to re-run the job keeping all input > mtz and model pdb and other parameters the same. (I simply click 'ReRun > job' on the menu.) The only adjustments were (1) ensemble IDENT from 90% to > 95%; (2) No. of copies to search for from 2 to 1. This time the programme > warns of significant twinning: > > > > > > -------- > > TWINNING > > -------- > > > > tNCS/Twin Detection Table > > ------------------------- > > -Second Moments- --P-values-- > > Centric Acentric untwinned twin > frac <5% > > Theoretical for untwinned 3.00 2.00 > > Theoretical for perfect twin 2.00 1.50 > > Initial (data as input) 0.00 1.35+/-0.013 0 0 > > After Anisotropy Correction 0.00 1.29+/-0.011 0 0 > > After Anisotropy and tNCS ---n/a--- > > > > P-value < 0.01 for <5% twinned is considered significant > > Resolution for Twin Analysis (85% I/SIGI > 3): 3.15A (HiRes= 2.67A) > > > > > ------------------------------------------------------------------------------------- > > Warning: Intensity moments suggest significant twinning (>5%). Tests > based on > > possible twin laws will be more definitive. > > > ------------------------------------------------------------------------------------- > > > > > > To my understanding Phaser first treats the data for twin / not twin > before it really puts in the model. Thus I am curious why the same set of > input data and model give different results in Twinning test? > > > > Thanks! > > > > Sam > > > > PhD Candidate, The Chinese University of Hong Kong > > ------ > Randy J. Read > Department of Haematology, University of Cambridge > Cambridge Institute for Medical Research Tel: + 44 1223 336500 > Wellcome Trust/MRC Building Fax: + 44 1223 336827 > Hills Road E-mail: [log in to unmask] > Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk > >