[ccp4bb]: TLS refinement

[Thomas Scott Reid <tsr@duke.edu>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***


I am a brand new user of ccp4.  Although weened on CNS, presentations at
the San Antonio ACA conference has convinced my to try CCP4 for some of my
refinement, especially TLS refinement in Refmac5.  I can get refmac to
refine my structure just fine, but the TLS part of the refinement doesn't
seem to work.

About my project: My crystals contain 6 molecules per asymmetric unit.
Each molecule is a heterodimer consisting of an alpha and beta subunit
(molecule 1 = chain A + chain B, mol2 = chain C + chain D, etc., etc.).
My current model has been fully refined in CNS with Rfree and Rcryst at
24.0 and 21.4%.  Waters and ligands are included in the model. My data set
is complete out to 2.4 A.  The average B factor is ~ 55 A^2. The protein
N-terminal domains have not been modeled, and some side chains have not
been modeled because they have no omit density (Is this important?).

I want to see what happens if I add TLS parameters to each of the six
molecules in the asymmetric unit.  I have not included water or ligands in
each TLS group (should I?).  The starting TLS input file is as follows:

TLS mol 1
RANGE 'A 60.' 'A 360.' all
RANGE 'B 20.' 'B 360.' all

TLS mol2
RANGE 'C 60.' 'C 360.' all
RANGE 'D 20.' 'D 360.' all

TLS mol3
RANGE 'E 60.' 'E 360.' all
RANGE 'F 20.' 'F 360.' all


etc., etc.,


I run refmac5 (5 cycles TLS refinement, 5 cycles restrained ML
refinement), and I see no error messages during the initial TLS refinement
(with B values fixed at 50 A^2).  Refmac spits out perfectly fine .pdb,
...tls and .lib files.  However, during TLS refinement, refmac always picks
the center of mass for each TLS group as 0,0,0 and assigns TLS values of 0
for all paramters.  The refined .tls file for the six groups is as
follows:

TLS mol 1
RANGE 'A 60.' 'A 360. 'all
RANGE 'B 20.' 'B 360. 'all
ORIGIN 0.000 0.000 0.000
T     0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
L     0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
S     0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000

TLS mol2
RANGE 'C 60.' 'C 360. 'all
RANGE 'D 20.' 'D 360. 'all
ORIGIN 0.000 0.000 0.000
T     0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
L     0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
S     0.0000 0.0000 0.0000 0.0000 0.00000.0000 0.0000 0.0000

etc., etc.,


What am I doing wrong?  Its probably something very simple, and related to
the fact that the PDB file comes straight out of CNS.  Any suggestions
would be appreciated.


Scott Reid
Dept. of Biochem
Duke University Medical Center
tsr@duke.edu