[Chimera-users] building a alpha helix and flexing it

Elaine Meng meng at cgl.ucsf.edu
Fri Nov 21 17:48:53 PST 2008

Hi Andy,
Any of the following are possible:
(1) warp the helix
(2) rotate bonds in the helix backbone to bend it
(3) start over and build it with different backbone angles
(4) just take a bent helix out of some existing PDB structure


(1) A structure can be warped using "Movement Mouse Mode" (under  
Tools... Movement) to move some parts while other parts are kept  
frozen.  If your helix was residues 1-10 in model 2, you could select  
only its first half, for example with the command:
select #2:1-5
Choose "Move selection" in the Movement Mouse Mode tool.  Then, moving  
the structure with the mouse will only move the selected part, and  
bonds to the other part will simply stretch or compress.  There is no  
attempt to give a physically meaningful result.  To move everything  
instead of just the selection, choose the "Normal" movement mouse mode.

(2) you can rotate bonds in the backbone of the helix to bend it. One  
way to make a bond rotatable is to Ctrl-doubleclick it and then click  
Rotate Bond in the little menu that appears.  This will bring up the  
Adjust Torsions dialog with that bond listed, and you can repeat the  
process to make several bonds rotatable.  The bonds can be rotated  
with the dials or by entering different angle values.  Once a bond is  
in the desired position, deactivate (freeze) it by clicking its entry  
in the Adjust Torsions dialog and choosing Deactivate from the menu  
that appears.  However, there is nothing to prevent you from putting  
atoms in bad positions clashing with each other, and nothing to tell  
you which torsion angles are reasonable.  Unlike option #1, however,  
the bond lengths and bond angles (the angles formed by 3 atoms) are  
kept reasonable.

For both #1 and #2, you can use Minimize Structure (under Tools...  
Structure Editing) afterward.  However, it might change the helix so  
that it doesn't fit the density as well.  If you do minimize, some  
important tips are:
(a) if possible, keep some atoms fixed (only let some of the atoms  
move during minimization) by selecting them and specifying "fixed  
atoms" as "selected," or selecting the OTHER atoms and specifying  
"fixed atoms" as "unselected"
(b) don't have any other structures around (close other models so they  
don't interfere)

(3) build the peptide again except specifying different backbone  
angles for some of the residues.  If addaa, you can only get certain  
combinations of phi/psi values, as shown in the man page:
If Build Structure, you can enter whatever phi/psi values you want.   
In the most recent daily builds of Chimera (from about the last week)  
this has been improved so that is much easier to specify a different  
phi and psi for each peptide bond.  However, the hard part is figuring  
out what values you actually want.   You could measure the values in a  
bent helix in some existing PDB file with the "angle" command
or the "Angles/Torsions" tool
However, if you already know of a structure that has a helix similar  
to what you want, you might as well go to the next option...

(4) just use a bent helix from some existing structure.  You could  
open that structure and delete all of its atoms that you don't want,  
or edit them out of the file before opening it.  If only its sequence  
is wrong (you want different sidechains), you can use the command  
or the Rotamers tool
to "mutate" the residues.

Probably none of these options are very convenient for your purposes,  
but they are all I can think of today.  Best,
Elaine C. Meng, Ph.D.                          meng at cgl.ucsf.edu
UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab
Department of Pharmaceutical Chemistry
University of California, San Francisco

On Nov 21, 2008, at 3:48 PM, Anindito Sen wrote:

> Hi Elaine
> In continuation of the process (mentioned below) I have generated  
> the helix that needs to be added. However it seems that the helix  
> needs to be flexed in some points to get a better fit. Can yo let me  
> know how do that?
> Thanks
> Andy
>> (1) With the "addaa" command you can build out from the C-term of a
>> protein, residue by residue.
>> <http://www.cgl.ucsf.edu/chimera/current/docs/UsersGuide/midas/addaa.html
>>  >
>> (2) With the "Build Structure" tool you can create a peptide that is
>> not attached to anything, then move it to a reasonable position and
>> combine it with the existing protein.
>> <http://www.cgl.ucsf.edu/chimera/current/docs/ContributedSoftware/editing/editing.html
>>  >

More information about the Chimera-users mailing list