VMD-L Mailing List

From: Charles Moad (cmoad_at_indiana.edu)
Date: Thu Apr 14 2005 - 20:08:24 CDT

        Go figure! 1di9 does work for me, but I do still see the issue in a
multiple frame molecule, 1dt7. I am guessing this issue is just related
to my frame setup, so thanks for all your help.

- Charlie

John Stone wrote:
> Charles,
> I just ran your script and it works perfectly as-written on my
> test machine here in the lab. I get a red line that starts at one
> of the CA atoms, and runs to the origin. You can verify this for
> yourself by creating a VDW rep with an atom selection of
> "name CA and index 1" (just to be paranoid), run your script,
> and you'll see that your red line goes right up to the CA atom.
> If there's doubt, you can draw another sphere at the origin using
> Tcl just to check, with "draw sphere { 0 0 0 }". I got this to work
> fine on my box here. Please re-test on your machine. If you still
> have problems getting the correct result, then it is possible that
> there's a problem with the Python build in your VMD version, or
> something else that's platform or machine specific.
>
> Thanks,
> John Stone
> vmd_at_ks.uiuc.edu
>
> On Wed, Apr 13, 2005 at 06:40:54PM -0500, Charles Moad wrote:
>
>>The scale is flat out wrong if I don't adjust it. I am expecting a line
>>from the center of the scene to the first alpha carbon. (code below)
>>It is close so I am guessing there needs to be some small adjustment to
>>the center/translation. The rotations are identical.
>>
>>Thanks for looking at this,
>> Charlie
>>
>>-------------------------------------------------------
>>import Molecule as mol
>>import AtomSel as sel
>>import vmdnumpy
>>import graphics as gl
>>import molecule
>>import trans
>>
>>m = mol.Molecule()
>>m.load('1di9', filetype='webpdb')
>>mid = int(m)
>>
>>cas = sel.AtomSel('name CA', mid)
>>cas.frame(0)
>>
>>cgo = molecule.load('graphics', 'test')
>># Move to pdb coord system
>>trans.set_center(cgo, trans.get_center(mid))
>>trans.set_scale(cgo, trans.get_scale(mid))
>>
>>x, y, z = cas.get('x', 'y', 'z')
>>gl.color(cgo, 'red')
>>gl.line(cgo, (0,0,0), (x[0], y[0], z[0]), width=3)
>>-------------------------------------------------------
>>
>>John Stone wrote:
>>
>>>Charles,
>>> Can you send me the script you're performing these commands within?
>>>I can probably be more helpful if I could see what you're actually doing
>>>beyond these two lines. The scale command you've got there may be causing
>>>troubles. When you say it does not appear correct, do you have any
>>>judgement as to whether the origin, scale, or both are being drawn incorrectly
>>>for your graphics object? Did you do anything else with your transforms
>>>before this code is executed, or after?
>>>
>>> John Stone
>>> vmd_at_ks.uiuc.edu
>>>
>>>On Sun, Apr 10, 2005 at 07:13:31PM -0500, Charles Moad wrote:
>>>
>>>
>>>>What is the correct way to draw a graphics object in a pdb's coordinate
>>>>system in python? I tried:
>>>>
>>>>trans.set_center(cgo, trans.get_center(mid))
>>>>trans.set_scale(cgo, trans.get_scale(mid))
>>>>
>>>>Where cgo is the id of the graphics object, and mid is the molecule's id.
>>>>
>>>>This does not appear correct.
>>>>
>>>>Thanks,
>>>> Charlie
>>>
>>>
>