[Chimera-users] Connect and align duplicated models

Joe Ping-Lin Hsiao phsiao at cs.unc.edu
Fri Oct 14 07:40:22 PDT 2011


It works!
Thanks a lot, Tom!

I also tested it on a Mac.
The trick is to download a "Daily Build" Chimera. In a "Current
Production Release", I would get warning message "no symmetric
molecules are selected" if I try to send a "sym" command.

I'll try it on Windows later.

Thanks again!
Joe


On Wed, Oct 12, 2011 at 6:13 PM, Tom Goddard <goddard at sonic.net> wrote:
> Hi Joe,
>
>  I suspect the sym command made the molecule copy in the wrong place because
> the two original copies of 3ghg (#0 and #1) do not have the same atom
> coordinates.  This causes the measure rotation command to give the wrong
> rotation parameters.  This would happen if your two original copies came
> from the Unit Cell dialog.  That dialog creates copies of the molecule with
> different atom coordinates.  Instead you want to start with two copies of
> 3ghg with identical atom coordinates -- for example just open it twice.
>
>    open 3ghg
>    open 3ghg
>
> 3ghg contains two copies of the coiled coil so delete one copy
>
>    del :.G-L,.Q-T
>
> Now move model #1 relative to model #0 as you like.  Then use measure
> rotation and the sym command.  I did this and it worked fine as shown in the
> attached image.  The reason this is different from using two copies of the
> molecule made by unit cell is rather technical.  The unit cell dialog makes
> copies of the molecules where the atoms have different x,y,z coordinates in
> the two copies.  The procedure I just described uses two molecule copies
> with the same atom coordinates but the models have different coordinate
> systems.  Moving one molecule relative to the other just changes the
> coordinate system of the moved molecule, not the actual atom x,y,z
> coordinate values.
>
>  The second problem you observed is that the molecule made by the sym
> command appears wrong -- as if it has all kinds of long bonds.  This could
> be a bug in the code that makes the molecule copy or a bug in your graphics
> driver.  I don't see that with Mac Chimera.  If you can make it happen
> again, then save a session and use menu Help / Report a Bug..., say the
> molecule drawing is wrong, and attach the session file.
>
>    Tom
>
>
>> Thanks, Elaine and Tom,
>>
>> Tom,
>>    I tried what you suggested but only generating one extra copy of
>> model #1. But sym command created a weird-looking model at the same
>> place where the original #1 is
>> (http://dl.dropbox.com/u/3757507/chinema3.png).
>>
>> Below is the output of my "measure rotation #0 #1":
>>
>> Position of single_3ghg_monomer.pdb #2 (#1) relative to
>> single_3ghg_monomer.pdb (#0) coordinates:
>>   Matrix rotation and translation
>>      0.79947388   0.29593455  -0.52274683  66.20071818
>>     -0.40522870  -0.37669835  -0.83299943 -113.45813726
>>     -0.44343118   0.87779331  -0.18123932 -31.01344468
>>   Axis   0.92445135  -0.04285935  -0.37888360
>>   Axis point   0.00000000 -54.63874726 -41.89926982
>>   Rotation angle (degrees) 112.28611242
>>   Shift along axis  77.81257053
>>
>> And the sym command I used is:
>>
>> sym #1 group h,77.81,112.286,2 axis 0.92445,-0.04286,-0.37888 center
>> 0,-54.6387,-41.89927 coord #0
>>
>>
>> Thanks,
>> Joe
>>
>> On Tue, Oct 11, 2011 at 6:42 PM, Tom Goddard<goddard at sonic.net>  wrote:
>>>
>>> Hi Joe,
>>>
>>>  Elaine's suggestion is the way to go.  If you have two copies of your
>>> molecule #0 and #1 and you want to add additional subunits use the
>>> "measure
>>> rotation" and the sym commands to do it.  Here's an example.  Notice I
>>> take
>>> the axis, axis point, rotation angle, and shift along axis from the
>>> measure
>>> rotation and use those in the "sym" command to create 4 additional copies
>>> using helical symmetry.  A single rotation and translation always
>>> specifies
>>> a helical symmetry (rotation about an axis and shift along that axis).
>>>  In
>>> the case where the shift along the axis is 0 you get cyclic symmetry.
>>>
>>>  You'll probably find the "sym" command documentation useful
>>>
>>>    http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/sym.html
>>>
>>>  Tom
>>>
>>>
>>> measure rotation #0 #1
>>>
>>> Reply log:
>>> Position of 1E6J (#1) relative to 1E6J (#0) coordinates:
>>>  Matrix rotation and translation
>>>     0.96949029  -0.23745478  -0.06085881  98.69560733
>>>     0.03663564   0.38584119  -0.92183752   7.88875940
>>>     0.24237657   0.89148292   0.38276860   7.78837497
>>>  Axis   0.97552284  -0.16313335   0.14745407
>>>  Axis point   0.00000000  17.01960113  14.23231084
>>>  Rotation angle (degrees)  68.34295730
>>>  Shift along axis  96.14132684
>>>
>>> sym #1 group h,96.14,68.34,5 axis 0.9755,-0.1631,0.14745 center
>>> 0,17.02,14.23 coord #0
>>>
>>>
>>>> Hi Joe,
>>>> Hmm, well, you could measure the transformation between your first and
>>>> second copies with "measure rotation."  Then, if that corresponds to
>>>> some
>>>> symmetry group or product of groups, you could generate multiple
>>>> additional
>>>> copies with a single "sym" command using the "group" option.
>>>>
>>>>
>>>>
>>>> <http://plato.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/measure.html#rotation>
>>>> <http://plato.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/sym.html>
>>>>
>>>> However, I'm not sure how easy or difficult it would be to figure out
>>>> the
>>>> necessary sym command from the information given in the Reply Log from
>>>> the
>>>> measure command, i.e. whether it is a simple helical relationship or
>>>> something more complicated.  Maybe a python script to repeatedly apply
>>>> the
>>>> transformation to additional copies would be more effective (beyond my
>>>> skill
>>>> set, sorry).
>>>>
>>>> I'm not the most knowledgable here on this particular topic; perhaps the
>>>> others will have clearer ideas of how to attack the problem, or a
>>>> script.
>>>>  Best,
>>>> Elaine
>>>> ----------
>>>> Elaine C. Meng, Ph.D.
>>>> UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab
>>>> Department of Pharmaceutical Chemistry
>>>> University of California, San Francisco
>>>>
>>>> On Oct 11, 2011, at 1:04 PM, Joe Ping-Lin Hsiao wrote:
>>>>
>>>>> Hi,
>>>>> I are trying to use Chimera to produce duplicates of one fibrinogen,
>>>>> rotate and move them, and concatenate together in a certain way.
>>>>> The fibrinogen looks like this
>>>>> (http://dl.dropbox.com/u/3757507/chimera1.png).
>>>>> I want to duplicate the model, and connect them in a chain
>>>>> (http://dl.dropbox.com/u/3757507/chimera2.png).
>>>>> The second image shows only two instances, but there will be more.
>>>>>
>>>>> Currently I do it by hand, including rotating and aligning them in
>>>>> correct z positions.
>>>>> But it take a lot of time to do it because the duplicates generated by
>>>>> using Tools/Higher-order structure/Unit Cell are not in the same
>>>>> plane. I have to look back and forth in different directions to make
>>>>> sure them are aligned.
>>>>> So I are wondering is there any way to automate this process, such as
>>>>> apply the previous transformation to a new duplicate?
>>>>>
>>>>> Thanks,
>>>>> Joe
>>>>
>>>> _______________________________________________
>>>> Chimera-users mailing list
>>>> Chimera-users at cgl.ucsf.edu
>>>> http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
>>>>
>>>
>>>
>
>
>




More information about the Chimera-users mailing list