Thanks, Sabine.   You are right.  There was a mistake in my cif file in the 
line of the peptide definition.  I corrected it and now it is fine.  

-- Jianghai




On Nov 23, 2011, at 11:10 AM, Sabine Schneider wrote:

> Hi Jianghai,
> 
> check if the D-Phe is defined as peptide in the cif file
> 
> like for instance in alanine:
> 
> data_comp_list
> loop_
> _chem_comp.id
> _chem_comp.three_letter_code
> _chem_comp.name
> _chem_comp.group
> _chem_comp.number_atoms_all
> _chem_comp.number_atoms_nh
> _chem_comp.desc_level
> ALA      ALA 'ALANINE                             ' L-peptide          10   5 
> .
> #
> 
> Sabine
> 
> 
> On 11/23/2011 05:04 PM, Jianghai Zhu wrote:
>> I can take the D-PHE cif out and rename it to DPH.  But the peptide bonds at 
>> the N and C terminus of DPH are broken in COOT since COOT doesn't consider 
>> DPH as an amino acid.  Any way to circumvent that?  Thanks.
>> 
>> -- Jianghai
>> 
>> 
>> 
>> 
>> On Nov 23, 2011, at 7:51 AM, Paul Emsley wrote:
>> 
>>> On 22/11/11 23:25, Jianghai Zhu wrote:
>>>> I am building a cyclic peptide with a D amino acid in the middle.  First, 
>>>> how can I connect the N and C termini of the peptide and make it a peptide 
>>>> bond?
>>> There is no way (in Coot) at the moment, sorry.  This is something I am 
>>> working on [1].  If you make a link between them, Coot will draw a link - 
>>> but that is not (all) that you want (AFAICS).
>>> 
>>>> Second, how do I make the PHE to a D amino acid?  I know there is a 
>>>> library of D-Phenylalanine in mon_lib.cif, but the code for that is PHE-D? 
>>>> I cannot use 5 letters for the residure, can I?
>>>> 
>>> As I suspect you know, several D amino acids have specific three-letter 
>>> codes, but D-PHE does not.  Using a PDB file, Coot will not detect that you 
>>> mean PHE-D and not PHE for the restraints.  If you use mmCIF/pdbx then I 
>>> think it will work.
>>> 
>>> Paul.
>>> 
>>> [1] this is related to carbohydrate refinement and involves using LINKR 
>>> information (but I am having trouble extracting that using the current mmdb 
>>> libs - I'll poke Eugene about it again).
>>> 
> 

Reply via email to