[buster-discuss] Refining multisubunit protein complex

[Pietro Roversi]

Hello James,

>  Does BUSTER-TNT only accept single
> letter upper-case chain names? 

It will accept A and a as distinct chain labels so no problem there

> If so, how can I encode the chain
> sequence and NCS information into the tnt sequence file?

Well you will only declare the "capital chain" sequences and refer the "small 
chain" sequences to those with CHAIN cards at the top of the file:

CHAIN A A
CHAIN B B
.........
CHAIN N N
CHAIN 1 A
CHAIN 2 B 
.......
CHAIN  N

So when you generate the seq files for each chain separately, declare two 
copies of each and give the capital and small letters to $BDG_home/bin/
buster/pirToSeq.pl (for example first chain A and second chain a; and so on); 
then cat together these seq files. Do not forget to add the SS bonds.

For the ligands I suggest you use the dictionaries available on HicUp; if they 
are covalently linked just define the linkages geometries (BOND, BCORREL and 
ANGLE) in the same dictionary file and declare the linkage and the ligand in 
the sequence file.


Then for example soft NCS restraints are trivial:

CLUSTER ANTERM RESIDUES 1 - 35 CHAINS A a
CLUSTER ACTERM RESIDUES 435 - 550 CHAINS A a
.....
CLUSTER Ff CHAINS F f

and so on
 
> Can I put the "A" and "a" chains into a single chain with a gap in the
> sequence? 

Yes you could but this will make NCS trickier to specify the NCS

If the resolution is not great the best results will be obtained alternating 
refinements of xyz only and Bonly, with the Bonly refinements using COMBINE B 
cards to restrict the shits in B factors to one shift per residue (Gwyndaf 
Evans' protocol):

COMBINE B A|1:*
COMBINE B A|2:*
.......
COMBINE B a|1:*
........
COMBINE B n|650:*

If you need help do not hesitate to get in touch

Ciao

Pietro


-- 
Pietro Roversi - Laboratory of Molecular Biophysics - Biochemistry Dept.
University of Oxford - South Parks Road - Oxford OX1 3QU - England, UK
Tel. 0044 (0)1865 275385 - Fax. 0044 (0)1865 275182
http://biop.ox.ac.uk/www/lea/website/index.htm