On Mon, 2021-06-14 at 10:42 -0400, Cheng Zhang wrote:
> I had this issue before and tried several ways to fix it. The best way that
> works for me is to edit the PDB and cif
> files of this unnatural amino acid for Coot or Phenix.
I am already furrowing my brow...
> The goal is to allow COOT and Phenix to recognize it as an 'amino acid', not
> a random HETATM,
Yes, the former would be better than the latter...
> so they can add peptide bonds and apply appropriate geometry restrictions in
> the real-space refinement automatically.
That's right.
> But I will be interested to learn better ways if there are any.
Let's see...
> Here is how I do it: 1. Make a PDB file with Phenix's elbow based on the
> chemical structure of this amino acid.
i) that's not what I would do, but then again one would not reasonably be
expecting Nigel to use pyrogen or acedrg.
ii) This would be a good point to add step 0: "Check that the amino acid is not
already in the monomer library"
If so, we can use skip much of what follows. You can use the text search
tool in Coot or the ligand search tool
at the PDBe:
https://www.ebi.ac.uk/pdbe-srv/pdbechem/
> I usually use MDL or SDL files.
If we're starting from scratch, that is reasonable.
> 2. Change the names of atoms in the PDB file. You have to name the main
> chain atoms of amino acids specifically so
> COOT can read them as 'amino acids'. Open a PDB file of any protein and you
> will see those specific names (C, O, N, C,
> CA, CB). 3. Use this PDB file to make a cif file.
That all seems a bit Heath Robinson, pyrogen will do that for you.
> In the cif file, you have to define this amino acid as an 'L-peptide' or
> 'D-peptide' in the ligand type line.
This is important.
> I attached a cif file of L-cysteine. Pay attention to the ligand definition
> ('L-peptide') and the names of atoms.
Yes - the name of the group, that is.
> 4. Open the PDB and cif files in Coot. Merge this amino acid into your
> structure and change its numbering.
Coot will deal with the numbering for you (hopefully sensibly).
> Now you should be able to do real space refinement on it as if it is a
> natural amino acid.
Yay!
> Also, there are already many unnatural amino acids in COOT or Phenix
> database. You just have to find the right 3-
> letter code.
That's right.
Paul.
########################################################################
To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list
hosted by www.jiscmail.ac.uk, terms & conditions are available at
https://www.jiscmail.ac.uk/policyandsecurity/
