Aligning a protein into a membrane
Elaine: Relating to your instructions of last 19 Nov (I printed your mail, then unfortunately I deleted it) how to align a protein into a membrane (for both I have valid pdb and valid Amber parameters) I am probably made elusive mistakes at the stage of saving new pdbs. I aligned the protein into the membrane, then (with protein selected and membrane not active) I saved the new pdb, getting two files, "0" for the membrane and "1" for the protein. In Amber LEaP both these pdb file could be loaded and combined, however save top/crd failed because something was wrong with the membrane file. In fact, if the original pdb for the membrane is used, top/crd can be saved, though the protein is not accurately aligned. Clearly, my plan is to remove superimpositions among molecules for the "whole" obtained from Amber. Thanks francesco pietra ____________________________________________________________________________________ Be a better pen pal. Text or chat with friends inside Yahoo! Mail. See how. http://overview.mail.yahoo.com/
Hi Francesco, You can always view the chimera-users archives to see old messages, http://www.cgl.ucsf.edu/pipermail/chimera-users/index.html Here is the message describing how you can remove the overlapping molecules BEFORE you write the PDB file from Chimera http://www.cgl.ucsf.edu/pipermail/chimera-users/2007-November/ 002010.html and the one that mentioned writing PDB http://www.cgl.ucsf.edu/pipermail/chimera-users/2007-November/ 002008.html I didn't say much about writing PDB, but to retain the spatial relationship you have generated in Chimera, you must choose the option to "Save relative to model" and save both models relative to the same number (if they are in 0 and 1, either save both relative to 0 or save both relative to 1, it doesn't matter which). http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/savemodel.html#pdb Elaine ----- Elaine C. Meng, Ph.D. meng@cgl.ucsf.edu UCSF Computer Graphics Lab and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html On Nov 21, 2007, at 9:24 AM, Francesco Pietra wrote:
Elaine: Relating to your instructions of last 19 Nov (I printed your mail, then unfortunately I deleted it) how to align a protein into a membrane (for both I have valid pdb and valid Amber parameters) I am probably made elusive mistakes at the stage of saving new pdbs.
I aligned the protein into the membrane, then (with protein selected and membrane not active) I saved the new pdb, getting two files, "0" for the membrane and "1" for the protein.
In Amber LEaP both these pdb file could be loaded and combined, however save top/crd failed because something was wrong with the membrane file. In fact, if the original pdb for the membrane is used, top/crd can be saved, though the protein is not accurately aligned.
Clearly, my plan is to remove superimpositions among molecules for the "whole" obtained from Amber. Thanks francesco pietra
Elaine: Most sorry that you had to repeat. I succeeded now, with still a problem. I.e, the membrane lipid (POPC) is hydrated at the head (this is how it can be built with the plugin). Those molecules of water are still in the space cleaned from lipid molecules, i.e. they superimpose the protein. I am not sure if such water should be removed before solvating the system, adding ions and MD. If it should be removed, is is another command? Before posting, I have now checked the saved file for the membrane. Probably there is a more serious problem: all "TER" records inserted between the POPC residues and between the WAT residues have been removed on saving (using "Save relative to model"). That will create problems to Amber, unless lack of TER records is accounted for by the CONECT records added (in the membrane pdb file from the plugin I had to add all mentioned TER records to have the file accepted by Amber) Thanks francesco In contrast, the "TER" record between the protein and the single WAT molecule within the pore was not removed. --- Elaine Meng <meng@cgl.ucsf.edu> wrote:
Hi Francesco, You can always view the chimera-users archives to see old messages, http://www.cgl.ucsf.edu/pipermail/chimera-users/index.html
Here is the message describing how you can remove the overlapping molecules BEFORE you write the PDB file from Chimera http://www.cgl.ucsf.edu/pipermail/chimera-users/2007-November/ 002010.html
and the one that mentioned writing PDB http://www.cgl.ucsf.edu/pipermail/chimera-users/2007-November/ 002008.html
I didn't say much about writing PDB, but to retain the spatial relationship you have generated in Chimera, you must choose the option to "Save relative to model" and save both models relative to the same number (if they are in 0 and 1, either save both relative to 0 or save both relative to 1, it doesn't matter which). http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/savemodel.html#pdb
Elaine ----- Elaine C. Meng, Ph.D. meng@cgl.ucsf.edu UCSF Computer Graphics Lab and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html
On Nov 21, 2007, at 9:24 AM, Francesco Pietra wrote:
Elaine: Relating to your instructions of last 19 Nov (I printed your mail, then unfortunately I deleted it) how to align a protein into a membrane (for both I have valid pdb and valid Amber parameters) I am probably made elusive mistakes at the stage of saving new pdbs.
I aligned the protein into the membrane, then (with protein selected and membrane not active) I saved the new pdb, getting two files, "0" for the membrane and "1" for the protein.
In Amber LEaP both these pdb file could be loaded and combined, however save top/crd failed because something was wrong with the membrane file. In fact, if the original pdb for the membrane is used, top/crd can be saved, though the protein is not accurately aligned.
Clearly, my plan is to remove superimpositions among molecules for the "whole" obtained from Amber. Thanks francesco pietra
____________________________________________________________________________________ Be a better pen pal. Text or chat with friends inside Yahoo! Mail. See how. http://overview.mail.yahoo.com/
On Nov 21, 2007, at 1:14 PM, Francesco Pietra wrote:
Elaine: Most sorry that you had to repeat. I succeeded now, with still a problem. I.e, the membrane lipid (POPC) is hydrated at the head (this is how it can be built with the plugin). Those molecules of water are still in the space cleaned from lipid molecules, i.e. they superimpose the protein.
I am not sure if such water should be removed before solvating the system, adding ions and MD. If it should be removed, is is another command?
You would just use the same approach as for the lipid molecules - find some way of selecting the residues you don't want (using a distance cutoff from the protein atoms) and then delete them. Try different selecting commands similar to what I sent before and see if they select what you want. It is not possible to give the exact command without knowing all the details of your system.
Before posting, I have now checked the saved file for the membrane. Probably there is a more serious problem: all "TER" records inserted between the POPC residues and between the WAT residues have been removed on saving (using "Save relative to model"). That will create problems to Amber, unless lack of TER records is accounted for by the CONECT records added (in the membrane pdb file from the plugin I had to add all mentioned TER records to have the file accepted by Amber)
It is technically correct to remove the TERs between HETATM residues, since HETATM residues in PDB format are not connected to each other unless there are CONECT records that say they are connected. However, AMBER may need the technically incorrect format (sounds like it does), and you may just have to put the TER lines in yourself or write a script to do that. Best Elaine
--- Elaine Meng <meng@cgl.ucsf.edu> wrote:
On Nov 21, 2007, at 1:14 PM, Francesco Pietra wrote:
Elaine: Most sorry that you had to repeat. I succeeded now, with still a problem. I.e, the membrane lipid (POPC) is hydrated at the head (this is how it can be built with the plugin). Those molecules of water are still in the space cleaned from lipid molecules, i.e. they superimpose the protein.
I am not sure if such water should be removed before solvating the system, adding ions and MD. If it should be removed, is is another command?
You would just use the same approach as for the lipid molecules - find some way of selecting the residues you don't want (using a distance cutoff from the protein atoms) and then delete them. Try different selecting commands similar to what I sent before and see if they select what you want. It is not possible to give the exact command without knowing all the details of your system.
I tried a lot of different combinations, without picking the right one, in order to remove water from only the zone from where POPC molecules are removed. I changed selections in Taskbar .. Select ... Zone, as if they were reflected to the command-line command or not. In all cases, it was only the command command-line command select #0 & #1 z<1.5 that worked. Of course, select solvent .. delete did not work, all WAT residues were removed. That is, I don't understand is how "select" of command "select #0 & #1 z<1.5" operates, i.e. why it only recognizes POP residues and not WAT residues in the membrane pdb. I tried to remove TER records, and even renaming WAT: in any case, only POPC were removed.
Before posting, I have now checked the saved file for the membrane. Probably there is a more serious problem: all "TER" records inserted between the POPC residues and between the WAT residues have been removed on saving (using "Save relative to model"). That will create problems to Amber, unless lack of TER records is accounted for by the CONECT records added (in the membrane pdb file from the plugin I had to add all mentioned TER records to have the file accepted by Amber)
It is technically correct to remove the TERs between HETATM residues, since HETATM residues in PDB format are not connected to each other unless there are CONECT records that say they are connected.
All atoms, for both POPC molecules and water molecules, are called ATOM. This was done by the plugin. (if this matters). Thanks francesco
However, AMBER may need the technically incorrect format (sounds like it does), and you may just have to put the TER lines in yourself or write a script to do that. Best Elaine
____________________________________________________________________________________ Get easy, one-click access to your favorites. Make Yahoo! your homepage. http://www.yahoo.com/r/hs
On Nov 22, 2007, at 3:08 AM, Francesco Pietra wrote:
I tried a lot of different combinations, without picking the right one, in order to remove water from only the zone from where POPC molecules are removed. I changed selections in Taskbar .. Select ... Zone, as if they were reflected to the command-line command or not. In all cases, it was only the command command-line command
select #0 & #1 z<1.5
that worked. Of course, select solvent .. delete did not work, all WAT residues were removed. That is, I don't understand is how "select" of command "select #0 & #1 z<1.5" operates, i.e. why it only recognizes POP residues and not WAT residues in the membrane pdb. I tried to remove TER records, and even renaming WAT: in any case, only POPC were removed.
Hi Francesco, The example command above means: select atoms that are (1) in model 0 AND (2) that are in residues with any atom within 1.5 angstroms of model 1. Perhaps your POPC molecules are in model 0 and your water molecules are not. Hopefully your protein is model 1. You could try ":wat" instead of "#0" but I recommend spending a little time to understand command-line atom specification, which will allow to you generalize to your different situations. There are some examples on the second page of the PDF "quick reference" http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/quickref.pdf and this is the User's Guide page (see especially the parts about zones and combinations): http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/ frameatom_spec.html
All atoms, for both POPC molecules and water molecules, are called ATOM. This was done by the plugin. (if this matters). Thanks francesco
I am not sure what you mean by "plugin" (VMD plugin?) but I believe they are supposed to be HETATM in the file written out by Chimera. Maybe they are only HETATM when combined with the protein in a single file. Either way, it may still be necessary for you to edit the file depending on what AMBER needs, sorry. Best, Elaine ----- Elaine C. Meng, Ph.D. meng@cgl.ucsf.edu UCSF Computer Graphics Lab and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html
--- Elaine Meng <meng@cgl.ucsf.edu> wrote:
On Nov 22, 2007, at 3:08 AM, Francesco Pietra wrote:
I tried a lot of different combinations, without picking the right one, in order to remove water from only the zone from where POPC molecules are removed. I changed selections in Taskbar .. Select ... Zone, as if they were reflected to the command-line command or not. In all cases, it was only the command command-line command
select #0 & #1 z<1.5
that worked. Of course, select solvent .. delete did not work, all WAT residues were removed. That is, I don't understand is how "select" of command "select #0 & #1 z<1.5" operates, i.e. why it only recognizes POP residues and not WAT residues in the membrane pdb. I tried to remove TER records, and even renaming WAT: in any case, only POPC were removed.
Hi Francesco, The example command above means: select atoms that are (1) in model 0 AND (2) that are in residues with any atom within 1.5 angstroms of model 1. Perhaps your POPC molecules are in model 0 and your water molecules are not.
As I wrote, the plugin furnishes the membrane with the lipid molecules hydrated at the polar head. The pdb (all atoms ATOM) is a series of POPC residues alternated with series of WAT residues (as I renamed them, because they call them TIP3). I even exchanged the names, calling POP the water residues and WAT the lipid residues: still the water in the area freed from the lipid. This is why I asked what is "select" addressing.
Hopefully your protein is model 1.
Absolutely it is.
You could try ":wat" instead of "#0" but I recommend spending a little time to understand command-line atom specification, which will allow to you generalize to your different situations. There are some examples on the second page of the PDF "quick reference" http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/quickref.pdf
and this is the User's Guide page (see especially the parts about zones and combinations): http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/ frameatom_spec.html
I'll try also all that. I am replying just to assure you that - although naive and primitive - before asking for help I am reflecting.
All atoms, for both POPC molecules and water molecules, are called ATOM. This was done by the plugin. (if this matters). Thanks francesco
I am not sure what you mean by "plugin" (VMD plugin?)
Yes, VMD, the only one I am aware of.
but I believe they are supposed to be HETATM in the file written out by Chimera.
Yes, they are renamed HETATM by Chimera.
Maybe they are only HETATM when combined with the protein in a single file. Either way, it may still be necessary for you to edit the file depending on what AMBER needs, sorry.
Having to use more than one "house" in this case, I am becoming more patient than usual. Luckily, it is easy to write Python scripts for modifying the files. I came to Python just for that and it was an extraordinary encounter. Thanks francesco
Best, Elaine ----- Elaine C. Meng, Ph.D. meng@cgl.ucsf.edu UCSF Computer Graphics Lab and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html
____________________________________________________________________________________ Be a better pen pal. Text or chat with friends inside Yahoo! Mail. See how. http://overview.mail.yahoo.com/
Elaine: select :wat & #1 z<1.5 delete sel followed by select :popc & #1 z<1.5 delete sel work. That is, residue WAT is indeed in the membrane file, though it is not seen there by the command select #0 & #1 z<1.5 It is seen as solvent. In fact, as I told some emails ago, the Taskbar command Select Structure Solvent allowed to remove all WATs. On reading what you suggested, my understanding is that those three records for water are seen as solvent no matter how the residue is named. So that my scrambling of names failed to perform. Thanks francesco --- Elaine Meng <meng@cgl.ucsf.edu> wrote:
On Nov 22, 2007, at 3:08 AM, Francesco Pietra wrote:
I tried a lot of different combinations, without picking the right one, in order to remove water from only the zone from where POPC molecules are removed. I changed selections in Taskbar .. Select ... Zone, as if they were reflected to the command-line command or not. In all cases, it was only the command command-line command
select #0 & #1 z<1.5
that worked. Of course, select solvent .. delete did not work, all WAT residues were removed. That is, I don't understand is how "select" of command "select #0 & #1 z<1.5" operates, i.e. why it only recognizes POP residues and not WAT residues in the membrane pdb. I tried to remove TER records, and even renaming WAT: in any case, only POPC were removed.
Hi Francesco, The example command above means: select atoms that are (1) in model 0 AND (2) that are in residues with any atom within 1.5 angstroms of model 1. Perhaps your POPC molecules are in model 0 and your water molecules are not. Hopefully your protein is model 1. You could try ":wat" instead of "#0" but I recommend spending a little time to understand command-line atom specification, which will allow to you generalize to your different situations. There are some examples on the second page of the PDF "quick reference" http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/quickref.pdf
and this is the User's Guide page (see especially the parts about zones and combinations): http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/ frameatom_spec.html
All atoms, for both POPC molecules and water molecules, are called ATOM. This was done by the plugin. (if this matters). Thanks francesco
I am not sure what you mean by "plugin" (VMD plugin?) but I believe they are supposed to be HETATM in the file written out by Chimera. Maybe they are only HETATM when combined with the protein in a single file. Either way, it may still be necessary for you to edit the file depending on what AMBER needs, sorry. Best, Elaine ----- Elaine C. Meng, Ph.D. meng@cgl.ucsf.edu UCSF Computer Graphics Lab and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html
____________________________________________________________________________________ Get easy, one-click access to your favorites. Make Yahoo! your homepage. http://www.yahoo.com/r/hs
participants (2)
-
Elaine Meng -
Francesco Pietra