[gmx-users] Re: Re: Re: Re: g_bundle problem

[Justin A. Lemkul]

Stefan Hoorman wrote:
> 
> 
>     Stefan Hoorman wrote:
> 
>     <snip>
> 
>      > I have tried using g_sgangle, but the problem is that it does not
>      > calculate the principal axis of the helix. Since g_sgangle allows
>     me to
>      > choose at maximum 3 atoms per analysis, it is quite difficult to
>     perform
>      > such an analysis if one is to do this for 4 or five different systems
>      > and dimers.
>      > So, in this case, since g_bundle asks me to select a group of top
>     and a
>      > group of bottom atoms, I would need to create an index group with my
>      > first group (top) being the top residues of both my helices and the
>      > second group being the bottom residues of my helices. Or would I
>     create
> 
>     Per the documentation, that sounds right.
> 
>     "The program reads two index groups and divides both of them in -na
>     parts. The
>     centers of mass of these parts define the tops and bottoms of the axes."
> 
>     -Justin
> 
>      > a first group for my first helix and a second group for my second
>     helix?
>      >
> 
> 
> As I waited for the answer I ran some tests using: In a first analysis 
> the top residues for both helices in one group and the bottom residues 
> of both helices in another group and set the "-na" flag in g_bundle as "2".
> In a second analysis I used in my first helix a the first group and the 
> second helix as the second group and set the "-na" flag as "4".
> Both analysis were made using alpha carbons only.
> I had already measured the interhelical tilt between these helices with 
> another program called qhelix. The problem is that it gives me loads of 
> problems with the sampling, since qhelix only measures each frame at a 
> time and I have to convert each frame into a pdb file (etc), but anyway, 
> it gave me interhelical tilt values that allowed me to "see" a trend there.
> When I compared the results between both methods described above, the 
> second one gave similar results to the ones I got from qhelix.
> I wrote this because maybe I understood wrong your last email, the 
> phrase "Per the documentation, that sounds right." refered to the 
> sentece above or below your statement?
> Thank you
> 

Sorry, my embedded comment must not have been as clear as it looked in my head 
:)  The reference I made was to your first statement.  As worded, it appears 
that groups with residues defining "top" and "bottom" should be created, not 
differently helices independently.  If that does not gel with reality, then 
maybe the wording of the documentation is not quite right; otherwise you'll have 
to look into the code to see how it's doing the calculations.

-Justin

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================