VMD-L Mailing List

From: Dan Wright (dtwright_at_uiuc.edu)
Date: Mon Aug 01 2005 - 15:19:03 CDT

Samuel Flores wrote:
> Hmm..
>
> Well in principle this is fine, but I need to automate the process. Can
> this command be issued through tcl?

In priciple, yes, but it's neither documented nor supported :) This is a real
hack. Please don't expect it to work on any future versions of except by mere
accident. However, I just tried it and it seems to work OK. I'm posting this
to the list because I've gotten a couple inquiries about being able to
automate alignments from other people as well.

step 1) start the multiple alignment GUI from Extensions->Analysis->Multiple
Alignment . You don't actually need to use it, but it has to be running to
initialize data structures etc.

step 2) Run the alignment with a command like

"::MultiSeq::multiSeqDoAlign [atomselect 2 all] [atomselect 3 all] [atomselect
4 all]"

This will run the alignment and save the transformed coordinates internally.
They will not be applied to the open molecules yet (see below). The
atomselects provide the atomic coordinates to use in the alignment.

step 3) run "::MultiSeq::applyViewpointHack" to apply the transformed
coordinates to the loaded molecules.

You can then write PDBs or whatever you would like. These commands should be
scriptable and repeatedly callable, as long as the multiple alignment GUI is open.

There are a couple things to note:

1) this will (probably) break the multiple alignment GUI. Don't try to use it
after using these commands, I don't know what will happen.
2) As previously mentioned, this is not supported at all. Nor will it be.
This message is strictly informational. And will probably self-destruct after
you read it.

Dan

> Sam
>
>
> -----Original Message-----
> From: owner-vmd-l_at_ks.uiuc.edu [mailto:owner-vmd-l_at_ks.uiuc.edu] On Behalf Of
> Dan Wright
> Sent: Monday, August 01, 2005 9:57 AM
> To: Samuel Flores
> Cc: 'vmd'
> Subject: Re: vmd-l: weird atomselect problem
>
> Sam,
>
> You could use the Multiple Alignment VMD plugin to do the alignment instead.
>
> It uses STAMP to perform the alignment, which is much more robust then the
> built-in "measure fit" command. You can then save the transformed atomic
> coordinates back out to new PDB files.
>
> See
>
> http://www.ks.uiuc.edu/Research/vmd/plugins/multiseq/
>
> for more information.
>
> Dan
>
> Samuel Flores wrote:
>
>>Hi All,
>>
>>I'm trying to align two proteins using the measure fit command. However
>>when I made the appropriate selection (e.g. set frag1 [atomselect "residue
>>70 to 80"] ) and tried to calculate the transformation matrix using
>
> measure
>
>>fit, it told me the two selections had a different number of atoms. This
>>was odd, because the two structures have perfect homology and the same
>>number of atoms. So I used writepdb to output the selection to a file and
>>found that the CE atom of one of the Methionine residues was missing, even
>>though the original file had it. The homologous, second structure had no
>>such problem. I then tried selecting only alpha carbons and had the same
>>problem. This time, an entire residue was inexplicably missing from one
>
> of
>
>>the two selections!
>>
>>Could someone tell me what might be wrong with the way I use atomselect?
>>The input files are perfectly fine.
>>
>>Sam
>>
>>-----Original Message-----
>>From: owner-vmd-l_at_ks.uiuc.edu [mailto:owner-vmd-l_at_ks.uiuc.edu] On Behalf
>
> Of
>
>>Samuel Flores
>>Sent: Sunday, July 31, 2005 3:59 PM
>>To: 'vmd'
>>Subject: vmd-l: rotate atomic coordinates?
>>
>>Cool, that works.
>>
>>Does anyone know how I can rotate the coordinates of a molecule? It is
>
> not
>
>>enough to move the molecule in the screen display, I need to rotate the
>>atomic coordinates themselves.
>>
>>
>>Sam
>>
>>-----Original Message-----
>>From: B.D Allen [mailto:B.D.Allen_at_newcastle.ac.uk]
>>Sent: Sunday, July 31, 2005 6:59 AM
>>To: Samuel Flores
>>Subject: Re: vmd-l: trig functions in vmd
>>
>>Hi Sam
>> The following code prints out the values of sine(x) for 0<x<2pi. You just
>>need to use the 'expr' function for math expressions.
>>
>>set pi 3.1415926535
>>for {set x 0} {$x < 10} {incr x} {
>> puts [expr sin(2.0*$pi*$i/10)]
>> }
>>
>>Try googling for 'tcl expr', there's tons of stuff out there.
>>A good start is http://tmml.sourceforge.net/doc/tcl/expr.html.
>>Hope this helps.
>>
>>Ben
>>
>>
>>>Howdy All,
>>>
>>>Can anyone tell me how to use trig functions in tcl? It's my
>>>understanding
>>>that I need mathfunc, but I can't find this library anywhere, although
>>>manuals for it are everywhere.
>>>
>>>Sam
>>>
>>>
>>>
>>
>>Ben Allen
>>Molecular Photonics Laboratory
>>School of Natural Sciences
>>Bedson Building
>>University Of Newcastle Upon Tyne
>>Newcastle Upon Tyne
>>NE1 7RU
>>
>>http://www.ncl.ac.uk/mpl
>>+44 (0)191 2226485
>>
>>
>>
>
> 

-- 
Dan Wright
(dtwright_at_uiuc.edu)
(http://www.scs.uiuc.edu/)
(UNIX Systems Administrator, School of Chemical Sciences, UIUC)
(333-1728)