From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Wed Dec 27 2006 - 13:21:27 CST

Hi Matt,
you'll need to deal with two separate issues to get what you want:
1) creating a proper topology for the PETE polymer
2) getting coordinates for the membrane of interest

Before getting started, you'll want to make sure you've read the namd
tutorial (including appendices) and topology file tutorials at
http://www.ks.uiuc.edu/Training/Tutorials/. Doing what you want to is
still somewhat complicated; we're currently working on things like
arbitrary polymer builders that will make it much easier. For now,
though, you'll still have to do some things by hand, which is where the
tutorials are useful.

Dealing with these in order:
1. Building up pretty much any polymer using VMD/psfgen is done just
like a polypeptide. If you look at the topology file entries for any of
the amino acids, you'll see that bonds are specified to -C and +N --
these denote bonds to the previous and next monomers in the polymer
chain. You need to first generate a topology file entry for your PETE
monomer, and then add the proper bonds to cause it to be connected in a
chain (and also make terminal patches, like NTER and CTER, to properly
cap the ends of your polymer). You can get the PETE monomer topology
file using a molefacture function: In the tkcon, enter
::Molefacture::write_topfile pete.top $sel
where $sel is an atomselect object containing your PETE monomer. If your
version of molefacture doesn't have this let me know and I'll send you
the latest alpha (which also has this command present in the file menu).
You'll need to assign proper atom types either in molefacture or by
editing the topology file, and then add the bonds necessary for
polymerization. Armed with this topology file, feeding a pdb file
containing several PETE monomers to psfgen will let you build a polymer.

Since we're talking about a pdb file with the proper coordinates, this
leads us neatly into
2. Getting the coordinates. You'll need to generate a pdb file
containing the coordinates for some section of the membrane you want.
Since I really don't know what a PETE membrane would look like, the best
I can do is recommend the following steps:
-Write a single pete monomer as a pdb from vmd
-Use your favorite text editor or filter to make a pdb file with many
copies of it, with consecutive residue numbering
-Load that composite pdb into vmd, and use either the mouse or movement
commands to move the residues into the appropriate places for a pete
membrane
-Write the new pdb file, and then run it through psfgen to get the
polymer of interest

Once you've got a single membrane layer generated, you can look at
making several chains with different size holes punched in them to
generate the different layers.

Peter

wolfinbm_at_uci.edu wrote:
> Hello
> I finally got a working file of PETE, unfortunately, I have pretty much no
> idea how to combine the PETE fragment into a large polymer (or even into a
> PETE dimer, for that matter). My immediate goal is to be able to put the
> PETE fragments together to produce a polymer. Then I want to make the
> first layer of the membrane with a hole in the middle. Finally, I will
> want to make many layers with an appropriate sized hole in each layer to
> create a membrane with a conical pore. Any advice on where to start here?
>
> Thanks & Best,
> Matt
>
>
>> Hi Matt,
>> Molefacture (and, for that matter, all plugins) does not yet save its
>> state along with th rest of the VMD save state function. It's something
>> we are currently working on, but it is difficult to combine the
>> modularity (and variety of origins) of the plugins with the machinery
>> necessary for a true save state. In the meantime, though, you should be
>> able to just save your molecule as an XBGF (using the molefacture file
>> menu) and load it again to restore your place, since molefacture's
>> internal state is almost all calculated on the fly from the molecule
>> being worked on.
>> Peter
>>
>> wolfinbm_at_uci.edu wrote:
>>
>>> Hello
>>> The problem I had was fixed; however, I am having a problem saving my
>>> work. I can use the "save state" function in VMD, but when I re-load
>>> it,
>>> nothing displays (even though the "D" button is on). Any suggestions?
>>>
>>> Thanks & Best,
>>> Matt Wolfinbarger
>>>
>>>
>>>
>>>> Hello
>>>> Thank You--the new version seems to work. But I've still got a long
>>>> way
>>>> to go on building the membrane...
>>>>
>>>>
>>>
>>>> Thanks & Best,
>>>> Matt Wolfinbarger
>>>>
>>>>
>>>
>>>> Ah yes, the infamous spaces-in-filenames issue; this had been causing
>>>>
>>>>
>>> problems on windows. This bug should have been fixed in the 1.8.5 plugin
>>> update -- did you download a fresh copy of the 1.8.5 installer? If so,
>>> please try the 1.8.6 alpha instead
>>>
>>>
>>>> (http://www.ks.uiuc.edu/Research/vmd/alpha/). Please let me know if you
>>>>
>>>>
>>> still have problems.
>>>
>>>
>>>> Peter
>>>>
>>>> wolfinbm_at_uci.edu wrote:
>>>>
>>>>
>>>>> Hello
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>> Hi Matt,
>>>>>> two notes:
>>>>>> -which version of vmd are you using? I'm somewhat puzzled by these
>>>>>> error
>>>>>>
>>>>>>
>>>>>>
>>>>> messages...
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>> I'm using version 1.8.5. Maybe the program, or at least the
>>>>>> molefacture
>>>>>> plugin, did not install correctly. I will try uninstalling VMD,
>>>>>>
>>>>>>
>>>>>>
>>>>> download >whatever that latest version is available, and re-install.
>>>>>
>>>>>
>>> If that >doesn't work I'll send another message to the list and give
>>> more
>>>
>>>
>>>>>> specifics on the error message.
>>>>>>
>>>>>>
>>>>>>
>>>>> I tried re-installing and I am still getting the error message. When
>>>>> I go
>>>>> to the menu function "Build" in molefacture, I use the menu function
>>>>> "new
>>>>> molecule from fragment." Here is the details of the error message:
>>>>>
>>>>> (BEGIN)
>>>>> wrong # args: should be "::Molefacture::new_mol_from_fragment
>>>>> fragpath"
>>>>>
>>>>>
>>> wrong # args: should be "::Molefacture::new_mol_from_fragment fragpath"
>>> while executing
>>>
>>>
>>>>> "::Molefacture::new_mol_from_fragment C:/Program Files/University of
>>>>>
>>>>>
>>> Illinois/VMD/plugins/noarch/tcl/molefacture1.1/lib/basemol/benzene.mfrag"
>>>
>>>
>>>>> (menu invoke)
>>>>> (END)
>>>>>
>>>>> Note that this applies for any fragment--benzene is just the one I
>>>>>
>>>>>
>>> happened to pick this time.
>>>
>>>
>>>>> Thank You,
>>>>> Matt
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>
>>>
>>>
>
>