Hello everyone,

Thanks Paul! This allowed me to use the real space refine function in COOT to properly fit NAD without the nicotinamide ring flying apart (solving the first part of the problem) but it did not solve the problem of NAD imploding after refmac.

In addition to shifting NN1 and NN7 to the right one space I had to shift several hydrogens (NH2/4/5/6 and AH2/8) to the left one space. I tried running refmac on the PDB file with the atom names in their shifted positions, in their original positions, renaming the NN1/7 atoms to N1/7 in both the pdb and cif files and putting NN1/7 in quotes ('NN1 ', 'NN7 '). The latter two cases failed outright. NAD in the output PDB in the former two cases was the same; the ligands came apart just like when I tried running refmac before I could properly fit the ligands.

Has anyone else had this problem with NAD? Any suggestions on what I should try next?

Thanks!
Larry



On Wed, May 12, 2010 at 12:09 AM, Paul Emsley <[log in to unmask]> wrote:
Larry Grant wrote:


I am attempting to fit NAD to a 2.1 A lactate dehydrogenase structure I've built and I'm running into several problems:

1. I loaded NAD from the COOT monomer library and was able to fit roughly 2/3 of it into the density using the real space refine zone function but the nicotinamide moiety flies apart as a result of the refinement. My work around for this was to load another ligand and roughly fit just the nicotinamide ring then cut and paste the coordinates from the second ligand into the file for the first. I then merged the ligand with the LDH structure using the merge molecules function in COOT. I fit NAD to four different chains this way. Attempting to refine the ligands after merging also results in the ring coming apart.

https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind0905&L=COOT&P=R6200&1=COOT&9=A&J=on&d=No+Match%3BMatch%3BMatches&z=4



2. I was hoping that a round of refmac would then refine the entire ligand and thereby neatly fit the crudely placed nicotinamide ring. When I run refmac it fits the first NAD fairly rigidly such that while it does fit the nicotinamide ring it knocks the adenine moiety out of place. The other three NAD's look like they imploded. The atoms are too close together resulting in a structure which looks nothing like NAD and which contains multiple carbons with greater than four bonds. I was advised to create a new ligand dictionary and attempt refmac refinement with it.

That should not be necessary.


I have created files using refmac in review mode, loading the NAD coordinates into Sketcher and running libcheck, the PRODRG server and downloading from the HIC-UP server. Running refmac with any of these libraries produces the exact same result.

The devil's in the detail.


Trying to use the real space refine zone on the one ligand which didn't come apart gives me the error message "No Restraints Found!".

If you are talking about Coot, then you need to make sure that the new dictionary you read in matches the comp_id of the residue/monomer you are trying to refine.



Paul.