[ccp4bb] problem with residue numbering and Coot

[Vellieux Frederic]

Hi folks,

I'm having a problem with Coot. Molecular Replacement solution to the 
phase problem. The MR model has one small stretch of residues missing, 
for which there is density in the maps (both normal and difference 
density maps). Hence I have built the model into this density. Except 
that there is a problem with the "standard" residue numbering scheme one 
has to use in this class of enzymes. The numbering jumps from residue 
number 103 to residue number 105. As a result, Coot won't accept that 
the residues be linked together and forces them apart. The initial 
refinement round has corrected this somewhat (but still not entirely, 
even though I had introduced a LINK record in the PDB file before 
refinement). So I don't seem to be able to use Coot to correct this.

Hence the question: is there any way to force Coot to accept that these 
two residues be connected together even though the numbering scheme is 
not sequential? I haven't found the correct option anywhere. And I 
really do not want to use a sequential numbering scheme for refinement 
and model building only, since it would means plenty of residue number 
editing (by hand) to re-introduce all of this weird residue numbering 
scheme, everywhere in the subunit.

Thanks,

Fred.