Kip PHENIX doesn't use the "L-peptide" field in the automatic linking of amino acids. It's based on 1. the sequence in the model 2. the bond distance cutoff (3.0 from memory) 3. the atom names in the residue (C, N, CA, O) Cheers Nigel On Wed, Nov 14, 2012 at 11:55 AM, Kip Guja wrote:

It may be a good idea to open your cif file in a text editor and check that the residue type is specified correctly -- in this case, it should be "L-peptide"

Here's a few lines for the alanine cif file as an example. Note where it says "L-peptide" -- I could be wrong but I believe this enables phenix to recognize that it should be linked (form bonds) with adjacent amino acid residues. Sometimes Elbow will create a cif file that says "Ligand" in this field, which may be related to the "missing bond" that you observe.

This previous discussion may also be helpful: http://www.phenix-online.org/**pipermail/phenixbb/2008-** January/011981.htmlhttp://www.phenix-online.org/pipermail/phenixbb/2008-January/011981.html

Example Alanine cif file: # 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 _chem_comp.initial_date _chem_comp.modified_date _chem_comp.source ALA ALA 'Alanine' L-peptide 10 5 .

On Wed, 14 Nov 2012 10:13:58 -0800 Nigel Moriarty wrote:

...

Agnes

It should link automatically if the criteria as listed in the previous post are followed. You can check it in the .geo file. I'm happy to take a look at it if you send me the model or even just the three residues involved on list.

Nigel

NB. Any files sent to me will be held in strictest confidence.

On Wed, Nov 14, 2012 at 10:07 AM, Agnes Thiane Pereira Machado < [email protected]> wrote:

...

Hi We are trying to refine a protein structure with a (covalently) modified residue. The colosest discussion to our problem is:

http://www.phenix-online.org/**pipermail/phenixbb/2010-** August/015629.htmlhttp://www.phenix-online.org/pipermail/phenixbb/2010-August/015629.html

We ran:

phenix.elbow --residue SSC ourmodel.pdb

And read the due cif file at phenix refinement. The residue by itself is OK, but we were

not able to perform the peptide bond between C and N. How to complete the .cif file such that it has due restraints for the peptide bond with this modified residue?

Agnes

______________________________**_________________ phenixbb mailing list [email protected] http://phenix-online.org/**mailman/listinfo/phenixbbhttp://phenix-online.org/mailman/listinfo/phenixbb

-- Nigel W. Moriarty Building 64R0246B, Physical Biosciences Division Lawrence Berkeley National Laboratory Berkeley, CA 94720-8235 Phone : 510-486-5709 Email : [email protected] Fax : 510-486-5909 Web : CCI.LBL.gov

______________________________**_________________ phenixbb mailing list [email protected] http://phenix-online.org/**mailman/listinfo/phenixbbhttp://phenix-online.org/mailman/listinfo/phenixbb

-- Nigel W. Moriarty Building 64R0246B, Physical Biosciences Division Lawrence Berkeley National Laboratory Berkeley, CA 94720-8235 Phone : 510-486-5709 Email : [email protected] Fax : 510-486-5909 Web : CCI.LBL.gov

It would be better to fix the model. The edits you mention also have a cutoff of 3 angstrom by default so you'd have to change it. You can increase the automatic cutoff distance using the phil parameter peptide_nucleotide_distance_cutoff but it's always better to have a model with residues within 3 angstroms. Nigel On Wed, Nov 14, 2012 at 12:25 PM, Kip Guja wrote:

Hi Nigel,

Thank you for the clarification, I appreciate it!

So in this specific case maybe one possible solution is to use geometry restraints to bring the atoms close enough to form the bond in question?

Something like this was previously described by Pavel here: http://www.phenix-online.org/**pipermail/phenixbb/2008-** January/011977.htmlhttp://www.phenix-online.org/pipermail/phenixbb/2008-January/011977.html

Kip

On Wed, 14 Nov 2012 12:10:23 -0800 Nigel Moriarty wrote:

...

Kip

PHENIX doesn't use the "L-peptide" field in the automatic linking of amino acids. It's based on

1. the sequence in the model 2. the bond distance cutoff (3.0 from memory) 3. the atom names in the residue (C, N, CA, O)

Cheers

Nigel

On Wed, Nov 14, 2012 at 11:55 AM, Kip Guja wrote:

It may be a good idea to open your cif file in a text

...

editor and check that the residue type is specified correctly -- in this case, it should be "L-peptide"

Here's a few lines for the alanine cif file as an example. Note where it says "L-peptide" -- I could be wrong but I believe this enables phenix to recognize that it should be linked (form bonds) with adjacent amino acid residues. Sometimes Elbow will create a cif file that says "Ligand" in this field, which may be related to the "missing bond" that you observe.

This previous discussion may also be helpful: http://www.phenix-online.org/****pipermail/phenixbb/2008-**http://www.phenix-online.org/**pipermail/phenixbb/2008-** January/011981.html<http://**www.phenix-online.org/** pipermail/phenixbb/2008-**January/011981.htmlhttp://www.phenix-online.org/pipermail/phenixbb/2008-January/011981.html

...

Example Alanine cif file: # 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 _chem_comp.initial_date _chem_comp.modified_date _chem_comp.source ALA ALA 'Alanine' L-peptide 10 5 .

On Wed, 14 Nov 2012 10:13:58 -0800 Nigel Moriarty wrote:

Agnes

...

It should link automatically if the criteria as listed in the previous post are followed. You can check it in the .geo file. I'm happy to take a look at it if you send me the model or even just the three residues involved on list.

Nigel

NB. Any files sent to me will be held in strictest confidence.

On Wed, Nov 14, 2012 at 10:07 AM, Agnes Thiane Pereira Machado < [email protected]> wrote:

Hi

...

We are trying to refine a protein structure with a (covalently) modified residue. The colosest discussion to our problem is:

http://www.phenix-online.org/****pipermail/phenixbb/2010-**http://www.phenix-online.org/**pipermail/phenixbb/2010-** August/015629.html<http://www.**phenix-online.org/pipermail/** phenixbb/2010-August/015629.**htmlhttp://www.phenix-online.org/pipermail/phenixbb/2010-August/015629.html

...

We ran:

phenix.elbow --residue SSC ourmodel.pdb

And read the due cif file at phenix refinement. The residue by itself is OK, but we were

not able to perform the peptide bond between C and N. How to complete the .cif file such that it has due restraints for the peptide bond with this modified residue?

Agnes

______________________________****_________________ phenixbb mailing list [email protected] http://phenix-online.org/****mailman/listinfo/phenixbbhttp://phenix-online.org/**mailman/listinfo/phenixbb http://phenix-online.org/mailman/listinfo/phenixbb

...

-- Nigel W. Moriarty Building 64R0246B, Physical Biosciences Division Lawrence Berkeley National Laboratory Berkeley, CA 94720-8235 Phone : 510-486-5709 Email : [email protected] Fax : 510-486-5909 Web : CCI.LBL.gov

______________________________****_________________ phenixbb mailing list [email protected] http://phenix-online.org/****mailman/listinfo/phenixbbhttp://phenix-online.org/**mailman/listinfo/phenixbb http://phenix-online.org/mailman/listinfo/phenixbb

...

-- Nigel W. Moriarty Building 64R0246B, Physical Biosciences Division Lawrence Berkeley National Laboratory Berkeley, CA 94720-8235 Phone : 510-486-5709 Email : [email protected] Fax : 510-486-5909 Web : CCI.LBL.gov

______________________________**_________________ phenixbb mailing list [email protected] http://phenix-online.org/**mailman/listinfo/phenixbbhttp://phenix-online.org/mailman/listinfo/phenixbb

-- Nigel W. Moriarty Building 64R0246B, Physical Biosciences Division Lawrence Berkeley National Laboratory Berkeley, CA 94720-8235 Phone : 510-486-5709 Email : [email protected] Fax : 510-486-5909 Web : CCI.LBL.gov