On Tue, Jun 19, 2007 at 09:08:02PM +0100, Janet Deane wrote:
> Dear all,
>
> I am building into a 3.5A map and am having difficulty with some loop
> regions. Is there a way in coot to use backbone libraries to maintain
> reasonable stereochemistry for these loops? If not, what's the best
> way to build reasonable loops at low resolution in a coot-centric
> world?
Dear Janet,
That is (in various forms) the most frequently asked question that
I have been asked recently.
I am sad to say, there is no way at present to do what you
want to do. That is to say, the Ramachandran plot is not part
of the geometry target function in real space refinement.
I understand that the fallback method of choice is to use the lego_ca
function in O to use library loops to build your missing fragment
and "improve" the backbone angles.
Not that it helps you right now, I'd like to add that Coot should do
better here and Kevin and I have plans.
Paul.
|