I was using X-PLOR at the time. I later switched to a PRESidue patch. I was not sure how many atoms must be added to a residue before a modified amino acid becomes a linked amino acid. The PDB seems a bit adverse to non-standard links. For example, why is NADP not two residues with a LINK? There are no well-defined rules, so I thought that the big nonstandard residue would better fit the usual pattern of modified residues. Before submitting, I switched to two residues because it made more sense to me.
However, a modified residue requires a MODRES entry, which was deemed "ugly", at least for the way that nucleic acids handled them. I think that many modified nucleic residues were converted to non-standard nucleic residues. I think it would be good to have some well-defined rules about what modifications produces a non-standard amino acid versus a linked residue. Marius Schmidt wrote: > What refinement program are you using? I assume cns. > Ihe idea with the non-standard residue works > very good, but the usual way is to PATCH the covalently > linked molecule to its respective covalently bound atom > in cns or xplor. > So, you do not need to alter your pdb-file. > Define a PRESI entry in topology file and set > parameters in param file (see how it is done > for standard patches in the existing top and param > files).
