[log in to unmask] wrote:
> I, and probably many coot users, would appreciate it very much if you
> could add a "create covalent link" option in coot instead of having us
> wade through 8 pages of instructions and hoping that we put exactly the
> right number of space characters in the link record.
>
Dear Herman (and for the Record),
As I am sure that you know, there are two issues here (which seem easily
conflated by a new-comer[1]).
One is the display of a LINK, the other is the use of the LINK in
refinement.
As for the former, after some reflection, I agree with you that it it
would be useful if you could go "click click" and add a LINK to the
molecule. I will add that to the specs for the next release. That will
deal with the esoteric documentation and correct number of spaces issues.
The latter is a more substantial issue it seems to me. Coot needs to be
told about the geometry of the link so that it can refine it. The
definition of the link bond lengths, angles, torsions and the estimated
standard deviations of those and any new chiral centres and atom
deletions are not in Coot's domain. The CCP4 program that deals with
that is libcheck and JLigand is the new interface to that. Of course we
will work together to make the interface of Coot/jligand/libcheck/refmac
more slick, but at the moment it does mean a deal of reading.
Regards,
Paul.
[1] which is why this is for the Record, if that wasn't already apparent.
|