On 20/01/15 15:32, ΕΜΜΑΝΟΥΗΛ ΣΑΡΕΙΔΑΚΗΣ wrote:
> Dear All,
>
> I wonder if you could help me with this. I am trying to do some regularisation/real space refinement of a ligand with Coot. The ligand is a pseudopeptide with two non-standard aminoacids, followed by several standard ones. I assume that to do this correctly, I would need a cif file that specifies the two non-standard aa's plus the links between these two and between second and third (which is a standard one).
What's the group? You need 'peptide'.
> When trying to generate a cif with a program such as eLBOW (using a pdb as the starting reference), each of the non-standard aa's gets a cif of its own, and the third, standard one, is ignored, meaning that there are no specified links between the first two (non-standard) and even less between second (non-standard) and third (standard).
Have you read the JLigand tutorial?
> Coot of course keeps each aa as it should be, but breaks up the peptide chain completely.
When you try to do what?
>
> Could you direct me how to built a cif that includes such bond specifications, starting from a reference pdb or otherwise? Is there an altogether different way of getting Coot to do what I want?
You can use Prodrg (via Prodrgify in Coot) but that is not recommended
(because it requires chemistry perception)
I have no reason to doubt that Elbow restraints for modified amino acids
would be just fine. I wouldn't do it that way, of course though.
Paul.
|