creating a peptide bond between two fragments
Hi Folks, I am trying to add a four amino acid segment to the N-terminus of a protein in the pdb. I can generate the four residue stretch and move its C-term close to the N-term of the structured protein. I can then select the C' and N atoms, which are about 1.0 Angstrom apart. At that point, I am stuck. Not sure how to add a bond and then create a proper structure. I've played around with Join Models, but no luck. Without a peptide bond recognized by Chimera, ribbons are discontinuous. Any ideas? Many thanks, glenn
Glenn L. Millhauser Department of Chemistry & Biochemistry UC Santa Cruz Santa Cruz, CA 95064 831 459 2176 voice 831 566 3337 cell 831 459 2935 fax 831 824 4645 google voice
glennm@ucsc.edu
http://chemistry.ucsc.edu/~glennm http://www.chemistry.ucsc.edu/faculty/millhauser.html
Hi Glenn,
You can combine the segments in Chimera into one PDB. What I usually do is go to Favorites->Model Panel (which lists all the segments open in your current session), then there is a button on the right labeled 'copy/combine'. You would want the combine functionality.
Check the following for the Model Panel window: http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/modelpanel.html
http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/modelpanel.htmlAlternatively, there's a command-line version of 'combine' (which I haven't used yet); details here: http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/combine.html
Just to add, I would recommend minimizing the structure after combining the segments. This can be done in Chimera with Tools->Structure Editing->Minimize Structure. Also, besides checking the distance between the C and N atoms, you would want to make sure they are in the trans configuration for peptide bonds. In my limited experience, the minimization doesn't always fix peptide bonds that start in the cis configuration.
http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/combine.htmlHope this helps, Tim
On Tue, Jun 7, 2011 at 5:30 PM, Glenn Millhauser glennm@ucsc.edu wrote:
Hi Folks, I am trying to add a four amino acid segment to the N-terminus of a protein in the pdb. I can generate the four residue stretch and move its C-term close to the N-term of the structured protein. I can then select the C' and N atoms, which are about 1.0 Angstrom apart. At that point, I am stuck. Not sure how to add a bond and then create a proper structure. I've played around with Join Models, but no luck. Without a peptide bond recognized by Chimera, ribbons are discontinuous. Any ideas? Many thanks, glenn
Glenn L. Millhauser Department of Chemistry & Biochemistry UC Santa Cruz Santa Cruz, CA 95064 831 459 2176 voice 831 566 3337 cell 831 459 2935 fax 831 824 4645 google voice
glennm@ucsc.edu
http://chemistry.ucsc.edu/~glennm http://www.chemistry.ucsc.edu/faculty/millhauser.html
Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
Hi Glenn, It may be tricky because Join Models requires some overlap of the input structures -- not spatially, as it will move the models for you, but an extra atom from each model that will be eliminated as it is replaced by the other model. It is done that way so you there are fewer geometrical parameters for the user to specify -- only one torsion and one bond length.
Here's what I did to attach 1zik chain A to the end of 1gcn chain A. First some setup commands (same stuff could be done with menus and dialogs):
open 1gcn open 1zik # display all atoms display # get rid of waters and 1zik chain B just to simplify system delete solvent delete :.b # hide ribbon ~ribbon rlabel
For replacement by Join Models, we need an -OXT at the C-term of 1gcn and at least one hydrogen on the N-term of 1zik. Generally if these are missing, they can be added with AddH, or the corresponding command:
addh
1gcn already had an OXT on Thr29, but addh was needed to put hydrogens on the terminal N of 1zik Arg1. For Join Models, I select one of those N-term hydrogens of 1zik and the OXT of 1gcn (could also be done with the mouse, Ctrl-click on one and Shift-Ctrl-click on the other):
select #1:1.a@h1#0:29.a@oxt
Then start Build Structure (under Tools... Structure Editing) and go to Join Models tab. It already has 180 as the torsion angle for the peptide bond that will be created, CA-C-N-CA, so I go ahead and click Apply.
Now the whole thing is a joined single model, but there is still an extra hydrogen on the former N-terminal nitrogen of 1zik (since that nitrogen used to be tetrahedral but should now be a planar amide). Remove those incorrectly placed hydrogens. You could select them with the mouse and use command "del sel" or delete by name as in this command:
delete :1.a@h2,h3
If you run addh again, it will correctly place a single amide hydrogen on that nitrogen:
addh
Verify that you get a single continuous ribbon:
ribbon # refresh residue labels ~rlabel rlabel
I see the numbering may be somewhat unexpected: the whole 1gcn part got negative numbers. We hope to make peptide joining less tricky in the future.
Yet another complication is that if you are lacking an OXT, but addh does not add it, it is because Chimera knows that residue is not the "real" C-terminus because it is looking at the SEQRES information in the input file. In that case, you could try removing the SEQRES information from the PDB file before opening it in Chimera and running addh.
I hope this helps (rather than discourages!), Elaine ----- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
On Jun 7, 2011, at 2:30 PM, Glenn Millhauser wrote:
Hi Folks, I am trying to add a four amino acid segment to the N-terminus of a protein in the pdb. I can generate the four residue stretch and move its C-term close to the N-term of the structured protein. I can then select the C' and N atoms, which are about 1.0 Angstrom apart. At that point, I am stuck. Not sure how to add a bond and then create a proper structure. I've played around with Join Models, but no luck. Without a peptide bond recognized by Chimera, ribbons are discontinuous. Any ideas? Many thanks, glenn
Thank you Elaine and Tim -- much appreciated. I will give this a try. One more question -- as opposed to linking two fragments, is it possible (and perhaps easier) to build residues one at a time on to the N-term of a protein from the pdb? all the best, glenn
On Jun 7, 2011, at 8:16 PM, Elaine Meng wrote:
Hi Glenn, It may be tricky because Join Models requires some overlap of the input structures -- not spatially, as it will move the models for you, but an extra atom from each model that will be eliminated as it is replaced by the other model. It is done that way so you there are fewer geometrical parameters for the user to specify -- only one torsion and one bond length.
Here's what I did to attach 1zik chain A to the end of 1gcn chain A. First some setup commands (same stuff could be done with menus and dialogs):
open 1gcn open 1zik # display all atoms display # get rid of waters and 1zik chain B just to simplify system delete solvent delete :.b # hide ribbon ~ribbon rlabel
For replacement by Join Models, we need an -OXT at the C-term of 1gcn and at least one hydrogen on the N-term of 1zik. Generally if these are missing, they can be added with AddH, or the corresponding command:
addh
1gcn already had an OXT on Thr29, but addh was needed to put hydrogens on the terminal N of 1zik Arg1. For Join Models, I select one of those N-term hydrogens of 1zik and the OXT of 1gcn (could also be done with the mouse, Ctrl-click on one and Shift-Ctrl-click on the other):
select #1:1.a@h1#0:29.a@oxt
Then start Build Structure (under Tools... Structure Editing) and go to Join Models tab. It already has 180 as the torsion angle for the peptide bond that will be created, CA-C-N-CA, so I go ahead and click Apply.
Now the whole thing is a joined single model, but there is still an extra hydrogen on the former N-terminal nitrogen of 1zik (since that nitrogen used to be tetrahedral but should now be a planar amide). Remove those incorrectly placed hydrogens. You could select them with the mouse and use command "del sel" or delete by name as in this command:
delete :1.a@h2,h3
If you run addh again, it will correctly place a single amide hydrogen on that nitrogen:
addh
Verify that you get a single continuous ribbon:
ribbon # refresh residue labels ~rlabel rlabel
I see the numbering may be somewhat unexpected: the whole 1gcn part got negative numbers. We hope to make peptide joining less tricky in the future.
Yet another complication is that if you are lacking an OXT, but addh does not add it, it is because Chimera knows that residue is not the "real" C-terminus because it is looking at the SEQRES information in the input file. In that case, you could try removing the SEQRES information from the PDB file before opening it in Chimera and running addh.
I hope this helps (rather than discourages!), Elaine
Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
On Jun 7, 2011, at 2:30 PM, Glenn Millhauser wrote:
Hi Folks, I am trying to add a four amino acid segment to the N-terminus of a protein in the pdb. I can generate the four residue stretch and move its C-term close to the N-term of the structured protein. I can then select the C' and N atoms, which are about 1.0 Angstrom apart. At that point, I am stuck. Not sure how to add a bond and then create a proper structure. I've played around with Join Models, but no luck. Without a peptide bond recognized by Chimera, ribbons are discontinuous. Any ideas? Many thanks, glenn
Glenn Millhauser Department of Chemistry & Biochemistry UC Santa Cruz Santa Cruz, CA 95064
831 459 2176 voice 831 459 2935 fax 831 566 3337 cell 831 824 4645 google voice
glennm@ucsc.edu
Hi Glenn, If you were building out from the C-term, the "addaa" command would probably be the easiest approach in Chimera. http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/addaa.html
However, for putting residues on the N-term or for joining two experimentally determined structures, Join Models (part of Build Structure) is generally the way to go. Start Structure (another part of Build Structure) allows creating peptides with specified phi,psi angles. http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/editing/editing.html#build
The alternative presented by Tim is positioning the models "manually" and then combining them into a single model with Chimera, then adding a bond. Personally, I usually find it too difficult to position the models correctly and would rather let Join Models take care of that part. Minimization has limited power in correcting the result. However, if the two structures have several atoms' overlap (say a few redundant residues), you could use the "match" command to superimpose the overlapping parts, then delete the redundant atoms, then combine into a single model and add the bond. That works pretty well if the match atoms are in approximately the same conformation. I've used that approach before we had Join Models, for example to splice an NMR-determined peptide structure on the end of an X-ray structure in which those residues were not resolved.
I hope this helps, Elaine ---------- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
On Jun 7, 2011, at 9:19 PM, Glenn Millhauser wrote:
Thank you Elaine and Tim -- much appreciated. I will give this a try. One more question -- as opposed to linking two fragments, is it possible (and perhaps easier) to build residues one at a time on to the N-term of a protein from the pdb? all the best, glenn
On Jun 7, 2011, at 8:16 PM, Elaine Meng wrote:
Hi Glenn, It may be tricky because Join Models requires some overlap of the input structures -- not spatially, as it will move the models for you, but an extra atom from each model that will be eliminated as it is replaced by the other model. It is done that way so you there are fewer geometrical parameters for the user to specify -- only one torsion and one bond length.
Here's what I did to attach 1zik chain A to the end of 1gcn chain A. First some setup commands (same stuff could be done with menus and dialogs):
open 1gcn open 1zik # display all atoms display # get rid of waters and 1zik chain B just to simplify system delete solvent delete :.b # hide ribbon ~ribbon rlabel
For replacement by Join Models, we need an -OXT at the C-term of 1gcn and at least one hydrogen on the N-term of 1zik. Generally if these are missing, they can be added with AddH, or the corresponding command:
addh
1gcn already had an OXT on Thr29, but addh was needed to put hydrogens on the terminal N of 1zik Arg1. For Join Models, I select one of those N-term hydrogens of 1zik and the OXT of 1gcn (could also be done with the mouse, Ctrl-click on one and Shift-Ctrl-click on the other):
select #1:1.a@h1#0:29.a@oxt
Then start Build Structure (under Tools... Structure Editing) and go to Join Models tab. It already has 180 as the torsion angle for the peptide bond that will be created, CA-C-N-CA, so I go ahead and click Apply.
Now the whole thing is a joined single model, but there is still an extra hydrogen on the former N-terminal nitrogen of 1zik (since that nitrogen used to be tetrahedral but should now be a planar amide). Remove those incorrectly placed hydrogens. You could select them with the mouse and use command "del sel" or delete by name as in this command:
delete :1.a@h2,h3
If you run addh again, it will correctly place a single amide hydrogen on that nitrogen:
addh
Verify that you get a single continuous ribbon:
ribbon # refresh residue labels ~rlabel rlabel
I see the numbering may be somewhat unexpected: the whole 1gcn part got negative numbers. We hope to make peptide joining less tricky in the future.
Yet another complication is that if you are lacking an OXT, but addh does not add it, it is because Chimera knows that residue is not the "real" C-terminus because it is looking at the SEQRES information in the input file. In that case, you could try removing the SEQRES information from the PDB file before opening it in Chimera and running addh.
I hope this helps (rather than discourages!), Elaine
Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
On Jun 7, 2011, at 2:30 PM, Glenn Millhauser wrote:
Hi Folks, I am trying to add a four amino acid segment to the N-terminus of a protein in the pdb. I can generate the four residue stretch and move its C-term close to the N-term of the structured protein. I can then select the C' and N atoms, which are about 1.0 Angstrom apart. At that point, I am stuck. Not sure how to add a bond and then create a proper structure. I've played around with Join Models, but no luck. Without a peptide bond recognized by Chimera, ribbons are discontinuous. Any ideas? Many thanks, glenn
Glenn Millhauser Department of Chemistry & Biochemistry UC Santa Cruz Santa Cruz, CA 95064
831 459 2176 voice 831 459 2935 fax 831 566 3337 cell 831 824 4645 google voice
glennm@ucsc.edu
http://chemistry.ucsc.edu/~glennm
Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
Terrific -- many thanks, Elaine. That does help. all the best, glenn
On Jun 8, 2011, at 9:12 AM, Elaine Meng wrote:
Hi Glenn, If you were building out from the C-term, the "addaa" command would probably be the easiest approach in Chimera. http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/addaa.html
However, for putting residues on the N-term or for joining two experimentally determined structures, Join Models (part of Build Structure) is generally the way to go. Start Structure (another part of Build Structure) allows creating peptides with specified phi,psi angles. http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/editing/editing.html#build
The alternative presented by Tim is positioning the models "manually" and then combining them into a single model with Chimera, then adding a bond. Personally, I usually find it too difficult to position the models correctly and would rather let Join Models take care of that part. Minimization has limited power in correcting the result. However, if the two structures have several atoms' overlap (say a few redundant residues), you could use the "match" command to superimpose the overlapping parts, then delete the redundant atoms, then combine into a single model and add the bond. That works pretty well if the match atoms are in approximately the same conformation. I've used that approach before we had Join Models, for example to splice an NMR-determined peptide structure on the end of an X-ray structure in which those residues were not resolved.
I hope this helps, Elaine
Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
On Jun 7, 2011, at 9:19 PM, Glenn Millhauser wrote:
Thank you Elaine and Tim -- much appreciated. I will give this a try. One more question -- as opposed to linking two fragments, is it possible (and perhaps easier) to build residues one at a time on to the N-term of a protein from the pdb? all the best, glenn
On Jun 7, 2011, at 8:16 PM, Elaine Meng wrote:
Hi Glenn, It may be tricky because Join Models requires some overlap of the input structures -- not spatially, as it will move the models for you, but an extra atom from each model that will be eliminated as it is replaced by the other model. It is done that way so you there are fewer geometrical parameters for the user to specify -- only one torsion and one bond length.
Here's what I did to attach 1zik chain A to the end of 1gcn chain A. First some setup commands (same stuff could be done with menus and dialogs):
open 1gcn open 1zik # display all atoms display # get rid of waters and 1zik chain B just to simplify system delete solvent delete :.b # hide ribbon ~ribbon rlabel
For replacement by Join Models, we need an -OXT at the C-term of 1gcn and at least one hydrogen on the N-term of 1zik. Generally if these are missing, they can be added with AddH, or the corresponding command:
addh
1gcn already had an OXT on Thr29, but addh was needed to put hydrogens on the terminal N of 1zik Arg1. For Join Models, I select one of those N-term hydrogens of 1zik and the OXT of 1gcn (could also be done with the mouse, Ctrl-click on one and Shift-Ctrl-click on the other):
select #1:1.a@h1#0:29.a@oxt
Then start Build Structure (under Tools... Structure Editing) and go to Join Models tab. It already has 180 as the torsion angle for the peptide bond that will be created, CA-C-N-CA, so I go ahead and click Apply.
Now the whole thing is a joined single model, but there is still an extra hydrogen on the former N-terminal nitrogen of 1zik (since that nitrogen used to be tetrahedral but should now be a planar amide). Remove those incorrectly placed hydrogens. You could select them with the mouse and use command "del sel" or delete by name as in this command:
delete :1.a@h2,h3
If you run addh again, it will correctly place a single amide hydrogen on that nitrogen:
addh
Verify that you get a single continuous ribbon:
ribbon # refresh residue labels ~rlabel rlabel
I see the numbering may be somewhat unexpected: the whole 1gcn part got negative numbers. We hope to make peptide joining less tricky in the future.
Yet another complication is that if you are lacking an OXT, but addh does not add it, it is because Chimera knows that residue is not the "real" C-terminus because it is looking at the SEQRES information in the input file. In that case, you could try removing the SEQRES information from the PDB file before opening it in Chimera and running addh.
I hope this helps (rather than discourages!), Elaine
Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
On Jun 7, 2011, at 2:30 PM, Glenn Millhauser wrote:
Hi Folks, I am trying to add a four amino acid segment to the N-terminus of a protein in the pdb. I can generate the four residue stretch and move its C-term close to the N-term of the structured protein. I can then select the C' and N atoms, which are about 1.0 Angstrom apart. At that point, I am stuck. Not sure how to add a bond and then create a proper structure. I've played around with Join Models, but no luck. Without a peptide bond recognized by Chimera, ribbons are discontinuous. Any ideas? Many thanks, glenn
Glenn Millhauser Department of Chemistry & Biochemistry UC Santa Cruz Santa Cruz, CA 95064
831 459 2176 voice 831 459 2935 fax 831 566 3337 cell 831 824 4645 google voice
glennm@ucsc.edu
http://chemistry.ucsc.edu/~glennm
Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
Glenn Millhauser Department of Chemistry & Biochemistry UC Santa Cruz Santa Cruz, CA 95064
831 459 2176 voice 831 459 2935 fax 831 566 3337 cell 831 824 4645 google voice
glennm@ucsc.edu
On Jun 8, 2011, at 9:12 AM, Elaine Meng wrote:
If you were building out from the C-term, the "addaa" command would probably be the easiest approach in Chimera. http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/addaa.html
However, for putting residues on the N-term or for joining two experimentally determined structures, Join Models (part of Build Structure) is generally the way to go. Start Structure (another part of Build Structure) allows creating peptides with specified phi,psi angles. <http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/editing/editing.htm...
As of tomorrow's daily build, addaa will allow adding the the N terminus as well.
--Eric
Eric Pettersen UCSF Computer Graphics Lab http://www.cgl.ucsf.edu
participants (4)
-
Elaine Meng -
Eric Pettersen -
Glenn Millhauser -
Tim Travers