If you have model coordinates for your CSA, I send those to the PRODRG
server and let it generate a REFMAC style dictionary.
You will need to make sure it is labelled as a peptide - cf the standatd
residue cif files to see how to do that..
Then you need to enter the LINKR record into the pdb file connevting the
CSA ands your CYS.
But if you run refmac from the GUI with the REVIEW restraints option
that should be done for you.. I thinjk the cif file generated should
contain both the CSA entry and an apppropriately named LINKR dictionary..
Eleanor
On 05/20/2011 09:59 PM, Geoffrey Feld wrote:
Greetings fellow Crystallographers,
I'm working on a structure at 1.8-A resolution that contains an acetone
crosslink between 2 cysteines (crosslink was incorporated by adding
1,3-dichloroacetone). I figured that the easiest way to model this is to
mutate one of the cysteines to S-acetonylcysteine (CSA in the PDB) and then
link it to the other cys. I've seen how to do this in COOT using the
mutate-by-overlap function; however, CSA is of course not in the CCP4
library that is installed on our system. I've built restraints for CSA using
phenix.elbow and tried importing the residue into COOT that way, still to no
avail. So I figured the way to go now is to import the cif file directly
into the LIBCHECK library in our system and then I should (in theory) be
able to use mutate-by-overlap to place the residue. However, this is where
I'm stuck. I can't seem to figure out the notation for importing the cif
file into LIBCHECK. I tried using FILE_CIF CSA.cif and I get an error
"reading title of input file." What am I doing wrong? Is there another
approach I should consider? Any help or advice would be greatly appreciated.
Cheers,
Geoff
