Re: [Chimera-users] Merge PDB Question
On the issue of using Chimera to write PDB files, would it be possible to change the format of the PDB output Chimera generates? I'm not too well-versed in the details of the pdb file format, but I ran into issues using Chimera-generated structures with a Blender plugin that used BioPython to handle parsing the files. Although I can't remember the details off the top of my head, it involved the use of multiple 'END' statements to separate models (in Chimera-generated files). It was easy enough to write a script to substitute 'MODEL', 'ENDMDL' pairs, but I'm sure it would be better implemented in whatever library handles the file IO in Chimera. Jonathan
Date: Thu, 28 Sep 2006 10:38:00 -0700 From: Elaine Meng <meng@cgl.ucsf.edu> Subject: Re: [Chimera-users] Merge PDB Question To: "Jared Godar" <jared.godar@vanderbilt.edu> Cc: chimera-users@cgl.ucsf.edu Message-ID: <4F99156D-A681-4EFB-AC6C-5A0C57D029E0@cgl.ucsf.edu> Content-Type: text/plain; charset="us-ascii"
Hi Jared, Currently you cannot bond different models, and the only way to "merge" is to put the atoms in the same file before opening it. There are actually other reasons your command wouldn't work:
- seems to me that there is no model 1, only 1.1, 1.2, 1.3 and 1.4, so the first part would specify zero atoms - the ampersand & stands for intersection (logical AND, not english and) and should be omitted, e.g. command: bond #1.1:504.b@c #1.1:505.b@n
What I've done in the past to handle similar cases (splicing stuff from an NMR structure onto a Xray structure) is to first put the tail (with a few extra residues to overlap with residues that are already present) into a file by itself. Then I open that new file, so it will be a new, separate model. I then match that model onto the original model using atoms in the overlapping part. Then, write out another PDB file of the tail relative to the untransformed original model. Using text-editing, remove the part that overlaps since it was just there for positioning purposes. FINALLY (text-editing again) make another PDB file that combines the new tail with the original protein. They may need to have the same chain ID as well. Sorry this is such a long process. If you think you can hand- position the tail sufficiently well you could skip the parts about having the extra atoms, using them for matching, and deleting them.
If you remove any intervening TER, the bond will already be there when you open the new file.
I hope this helps, Elaine
Hi Jonathan, Good point. This is actually on my to-do list and there are two different entries in our bug database requesting this (#1549 and #2642). I'll increase the priority of getting it done, but it probably won't get into this next release since we are freezing the code for testing this Monday. --Eric On Sep 28, 2006, at 2:54 PM, Jonathan Hilmer wrote:
On the issue of using Chimera to write PDB files, would it be possible to change the format of the PDB output Chimera generates? I'm not too well-versed in the details of the pdb file format, but I ran into issues using Chimera-generated structures with a Blender plugin that used BioPython to handle parsing the files. Although I can't remember the details off the top of my head, it involved the use of multiple 'END' statements to separate models (in Chimera-generated files). It was easy enough to write a script to substitute 'MODEL', 'ENDMDL' pairs, but I'm sure it would be better implemented in whatever library handles the file IO in Chimera.
Jonathan
Date: Thu, 28 Sep 2006 10:38:00 -0700 From: Elaine Meng <meng@cgl.ucsf.edu> Subject: Re: [Chimera-users] Merge PDB Question To: "Jared Godar" <jared.godar@vanderbilt.edu> Cc: chimera-users@cgl.ucsf.edu Message-ID: <4F99156D-A681-4EFB-AC6C-5A0C57D029E0@cgl.ucsf.edu> Content-Type: text/plain; charset="us-ascii"
Hi Jared, Currently you cannot bond different models, and the only way to "merge" is to put the atoms in the same file before opening it. There are actually other reasons your command wouldn't work:
- seems to me that there is no model 1, only 1.1, 1.2, 1.3 and 1.4, so the first part would specify zero atoms - the ampersand & stands for intersection (logical AND, not english and) and should be omitted, e.g. command: bond #1.1:504.b@c #1.1:505.b@n
What I've done in the past to handle similar cases (splicing stuff from an NMR structure onto a Xray structure) is to first put the tail (with a few extra residues to overlap with residues that are already present) into a file by itself. Then I open that new file, so it will be a new, separate model. I then match that model onto the original model using atoms in the overlapping part. Then, write out another PDB file of the tail relative to the untransformed original model. Using text-editing, remove the part that overlaps since it was just there for positioning purposes. FINALLY (text-editing again) make another PDB file that combines the new tail with the original protein. They may need to have the same chain ID as well. Sorry this is such a long process. If you think you can hand- position the tail sufficiently well you could skip the parts about having the extra atoms, using them for matching, and deleting them.
If you remove any intervening TER, the bond will already be there when you open the new file.
I hope this helps, Elaine
Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://www.cgl.ucsf.edu/mailman/listinfo/chimera-users
participants (2)
-
Eric Pettersen -
Jonathan Hilmer