Hi I think you have very good point here. Towards the end of refinement (when TLS refinement make more sense since the effects of positional coordinate errors on B values become smaller) phased refinement refinement makes less sense especially if phases are from density modification procedures (any procedure uses some assumptions/model and inevitable results are biased towards these assumptions/models, density modification is not an exception). In general it seems that when errors in model becomes comparable to the errors in the data then these data may not improve the model any more. In terms of phases when errors in phases become comparable with the errors in model phases then these phases may harm more than do any good. If probability distribution of phases would reflect their errors accurately then these phases could have been used longer in refinement. Unfortunately none of the generated phase probability distributions are that good (some programs produce better than others in generating these distributions). During phasing we do not have all information to generate these phase prob distributions accurately. There is an argument for using all data simultaneously but it poses different type of problems.
regards Garib On 11 Oct 2010, at 20:38, Tim Gruene wrote: > Hello, > > following this discussion, I was wondering how much sense it makes to use > TLS domains and experimental phases at the same time. > > I would have thought that TLS domains are better to be used when the model is > fairly complete at which state the quality of the experimental phases are > probably outrun by those of the model unless the experimental phases are > _really_ good. > > What is people's opinion about this? > > Cheers, Tim > > > On Mon, Oct 11, 2010 at 11:48:15AM -0700, Ethan Merritt wrote: >> On Monday, October 11, 2010 01:04:49 pm Christian Roth wrote: >>> Dear Ethan, >>> >>> thanks for your fast answer. The pdb file does the trick. I also did not >>> click >>> the reset B-factor box, which I thought is equivalent to the TLS+Biso pdb. >>> Is >>> that wrong? >> >> For what you are doing, it is correct to not click the box. >> I just wanted to confirm what were the steps leading to the observed failure. >> >>> Every program has his special requirements and I mixed it up with a >>> refinement >>> done some weeks before in phenix. >> >> So far as I know, there is still no provision in phenix for reading in >> existing TLS parameters. It always wants to regenerate them for itself. >> This make it very hard, if not impossible, to compare refinement protocols >> or validate previous refinements. >> >> Ethan >> >>> I'm sorry about that and thanks again for your help >>> >>> Christian >>> >>> Am Montag 11 Oktober 2010 19:50:10 schrieb Ethan Merritt: >>>> On Monday, October 11, 2010 11:01:10 am Christian Roth wrote: >>>>> Dear all, >>>>> >>>>> I want refine my model using TLS plus phase information from experimental >>>>> phases. >>>>> I used the TLSMD server to generate the initial tls parameters and >>>>> started Refmac >>>> >>>> I don't know exactly what has gone wrong. >>>> Let's collect some more information. >>>> >>>> - When you took the intial TLS parameters from TLSMD server, did you also >>>> take the corresponding modified PDB file? >>>> If not, that could well be the problem. >>>> >>>> - It looks from the output that you did not tick the "reset B factors to >>>> 20" box in ccp4i. Can you confirm that? >>>> >>>> Ethan >>>> >>>>> After an initial round of TLS refinement Refmac stops and and claims a >>>>> problem but nevertheless writes at the end "Task completed successfully" >>>>> and does not write any file despite it claims to have it done. I copied a >>>>> part of the log file at the end of the mail >>>>> Could anyone point me to the mistake i made probably? >>>>> >>>>> >>>>> Thanks in advance >>>>> >>>>> Christian >>>>> >>>>> >>>>> #CCP4I VERSION CCP4Interface 2.0.6 >>>>> #CCP4I SCRIPT LOG refmac5 >>>>> #CCP4I DATE 11 Oct 2010 17:37:30 >>>>> #CCP4I USER roth >>>>> #CCP4I PROJECT sharpoutputrefinement >>>>> #CCP4I JOB_ID 19 >>>>> #CCP4I SCRATCH /tmp >>>>> #CCP4I HOSTNAME bbzws339.bbz.uni-leipzig.de >>>>> #CCP4I PID 13018 >>>>> >>>>> <B><FONT COLOR="#FF0000"><!--SUMMARY_BEGIN--> >>>>> <html> <!-- CCP4 HTML LOGFILE --> >>>>> <hr> >>>>> <!--SUMMARY_END--></FONT></B> >>>>> <B><FONT COLOR="#FF0000"><!--SUMMARY_BEGIN--> >>>>> <pre> >>>>> >>>>> ############################################################### >>>>> ############################################################### >>>>> ############################################################### >>>>> ### CCP4 6.1: Refmac_5.5.0109 version 5.5.0109 : 13/10/09## >>>>> ############################################################### >>>>> User: unknown Run date: 11/10/2010 Run time: 17:37:30 >>>>> >>>>> >>>>> Please reference: Collaborative Computational Project, Number 4. 1994. >>>>> "The CCP4 Suite: Programs for Protein Crystallography". Acta Cryst. D50, >>>>> 760-763. >>>>> as well as any specific reference in the program write-up. >>>>> >>>>> <!--SUMMARY_END--></FONT></B> >>>>> $TEXT:Reference1: $$ comment $$ >>>>> "Refinement of Macromolecular Structures by the Maximum-Likelihood >>>>> Method:" G.N. Murshudov, A.A.Vagin and E.J.Dodson,(1997) >>>>> Acta Crystallogr. D53, 240-255 >>>>> EU Validation contract: BIO2CT-92-0524 >>>>> >>>>> $$ >>>>> $SUMMARY :Reference1: $$ Refmac: $$ >>>>> >>>>> :TEXT:Reference1: $$ >>>>> >>>>> Data line--- make check NONE >>>>> Data line--- make hydrogen ALL hout NO peptide NO >>>>> cispeptide NO ssbridge YES symmetry YES sugar YES >>>>> connectivity NO link NO >>>>> Data line--- refi type REST PHASE SCBL BBLU resi MLKF >>>>> meth CGMAT bref ISOT >>>>> Data line--- refi tlsc 10 >>>>> Data line--- ncyc 25 >>>>> Data line--- scal type SIMP LSSC ANISO EXPE >>>>> Data line--- solvent YES >>>>> Data line--- weight MATRIX 0.1 >>>>> Data line--- monitor MEDIUM torsion 10.0 distance 10.0 >>>>> angle 10.0 plane 10.0 chiral 10.0 bfactor 10.0 bsphere >>>>> 10.0 rbond 10.0 ncsr 10.0 >>>>> Data line--- labin FP=FP SIGFP=SIGFP HLA=HLA HLB=HLB HLC=HLC HLD=HLD >>>>> FREE=FreeR_flag >>>>> Data line--- labout FC=FC FWT=FWT PHIC=PHIC PHWT=PHWT DELFWT=DELFWT >>>>> PHDELWT=PHDELWT FOM=FOM PHCOMB=PHCOMB >>>>> Data line--- PNAME sharpoutputrefinement >>>>> Data line--- DNAME hrsync >>>>> Data line--- RSIZE 80 >>>>> Data line--- END >>>>> >>>>> OPENED INPUT MTZ FILE >>>>> Logical Name: HKLIN Filename: >>>>> /homes/roth/dlhdata/sharp/eden_unique.mtz >>>>> >>>>> >>>>> Spacegroup information obtained from library file: >>>>> Logical Name: SYMINFO Filename: >>>>> /software/linuxsoft/ccp4/ccp4-6.1.3/lib/data/syminfo.lib >>>>> >>>>> >>>>> **** Input and Default parameters# **** >>>>> >>>>> >>>>> Input coordinate file. Logical name - XYZIN actual file name - >>>>> /homes/roth/dlhdata/sharp/shrefmac9.pdb >>>>> Output coordinate file. Logical name - XYZOUT actual file name - >>>>> /homes/roth/dlhdata/sharp/shrefmac10.pdb >>>>> Input reflection file. Logical name - HKLIN actual file name - >>>>> /homes/roth/dlhdata/sharp/eden_unique.mtz >>>>> Output reflection file. Logical name - HKLOUT actual file name - >>>>> /homes/roth/dlhdata/sharp/eden_refmac3.mtz >>>>> >>>>> Cell from mtz : 61.560 69.500 133.350 90.000 90.000 >>>>> 90.000 Space group from mtz: number - 19; name - P 21 21 21 >>>>> >>>>> Refinement type : Restrained >>>>> >>>>> >>>>> ***TLS refinement cycle*** 2 >>>>> >>>>> 1 >>>>> 0.1153065 0.1254572 0.1168760 1.2506233E-02 >>>>> 2.1331017E-03 1.5424690E-04 >>>>> -1.4598943E-02 -3.3578421E-03 -1.7373050E-02 -3.3114674E-03 >>>>> -1.3628958E-02 -2.4346421E-03 >>>>> -8.4016532E-02 4.1347187E-02 -0.1191208 -0.1013065 -0.1222848 >>>>> 0.1371335 -0.2420141 5.1373582E-02 >>>>> -0.3787168 >>>>> Problem >>>>> xyz 1107 48.64835 38.25826 83.76399 -34.85817 >>>>> 8.685342 -6.930134 -12.26284 -18.79212 3.157892 >>>>> 19.69073 -34.85817 0.6871372 23.66340 -0.3420097 >>>>> -0.5949482 0.7273692 >>>>> ************************************************************************* >>>>> ** * Information from CCP4Interface script >>>>> ************************************************************************* >>>>> ** Writing final coordinates (XYZOUT) to >>>>> /homes/roth/dlhdata/sharp/shrefmac10.pdb >>>>> ************************************************************************* >>>>> ** >>>>> >>>>> ************************************************************************* >>>>> ** * Information from CCP4Interface script >>>>> ************************************************************************* >>>>> ** Writing final phases (HKLOUT) to >>>>> /homes/roth/dlhdata/sharp/eden_refmac3.mtz >>>>> ************************************************************************* >>>>> ** >>>>> >>>>> ************************************************************************* >>>>> ** * Information from CCP4Interface script >>>>> ************************************************************************* >>>>> ** Writing final TLS (TLSOUT) to >>>>> /homes/roth/dlhdata/sharp/dlhtls_refmac1.tls >>>>> ************************************************************************* >>>>> ** >>>>> >>>>> >>>>> #CCP4I TERMINATION STATUS 1 >>>>> #CCP4I TERMINATION TIME 11 Oct 2010 17:42:38 >>>>> #CCP4I TERMINATION OUTPUT_FILES >>>>> /homes/roth/dlhdata/sharp/sharpoutputrefinement_19_hrsync.refmac.cif >>>>> sharpoutputrefinement >>>>> #CCP4I MESSAGE Task completed successfully >>>> >>> >> >> -- >> Ethan A Merritt >> Biomolecular Structure Center, K-428 Health Sciences Bldg >> University of Washington, Seattle 98195-7742 > > -- > -- > Tim Gruene > Institut fuer anorganische Chemie > Tammannstr. 4 > D-37077 Goettingen > > phone: +49 (0)551 39 22149 > > GPG Key ID = A46BEE1A >
