On 01/27/2016 07:17 AM, DOOLAN, AISHLING wrote:
> This is my first entry on the CCP4bb. I am working on depositing a structure containing cytochrome c to the PDB.
>
> I used the 3tyi pdb file as a template when running pdbset and then Phaser Molecular replacement in CCP4. I carried out multiple rounds of refinement in Phenix.
>
> During deposition the heme group of cytochrome c was flagged as having no matches in the CCD, the ligand dictionary.
>
> In one chain (A) there are two additional protons on the heme that shouldn't be there. Any ideas as to how this happened and/or how to approach fixing it?
>
> In chain B I can't see any problems with the heme and can only guess it was flagged because it's geometry is slightly different to the one in the CCD. Is this a problem?
>
>
> <https://www.ebi.ac.uk/pdbe/entry/pdb/3tyi/>
>

Hi Aishling,
The last time I deposited a c-type heme there did seem to be some problems with the parameters,
both in CCP4 and the pdb, which I described here:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind1408&L=ccp4bb&F=&S=&P=116521

HOWEVER you say the heme has no matches in the ligand dictionary.
I think that match is made based on the 3-letter residue code for the
heme, not on geometry. If you had a glucose molecule and called it HEM or HEC
it would be recognized as such (but there would be a lot of missing atoms and
unrecognized atoms.
So what is the 3-letter code for heme in your deposition?
If HEM as indicated in your subject, then I don't understand it.
Maybe someone more familiar with the deposition process can answer.
Or send us the exact error message- maybe it is not a failure to
recognize the HEM but rather its atoms?

Cytochrome c structures usually call the heme HEC or HEM.
In principle the HEC.cif should stretch out the double bond in
the vinyl groups to single, whereas with HEM you wold have to do that with an edit.
In either case you need LINK records, not only for the axial ligand Met
and His, but also for the two cysteins that bond to the vinyl group.
I notice 3tyi uses HEM and lacks LINK records for the cys bonds.
eab