Re: [Jmol-users] Molecular view (CIF)

[<sw...@pu...>]

Quoting Dean Johnston <djo...@gm...>:

Thanks for this Dean - both scripts work nicely

I hadnt tried the extended load {333 777 1} before -
starting with it, I can now produce a correct 'molecular view' also.

Cheers

Simon

>
> Simon,
>
> Here's what I've used.  It obviously only works for molecular systems and
> can be somewhat slow for very large molecules/unit cells.  Bob can probably
> figure out a more efficient way to do this, but it works for me.  In these
> examples I've deleted any molecules that I don't want to see - you may want
> to just hide/display molecules instead.
>
> (1) I think this is equivalent to what Bob had suggested before - similar to
> Mercury's "include atoms in molecules where *Any* atom fits".
>
> load xxx.cif {333 777 1}; delete not within(molecule, unitcell);
>
>
> (2) Or (more complicated) - I think this is similar to Mercury's "include
> atoms in molecules whose *Centroids* fit".
>
> load xxx.cif {333 777 1};
> delete not within(molecule, unitcell);
> n = getProperty("moleculeInfo").size;
> for (var i=1; i<=n; i++)
>     {
>         select molecule=i;
>         if (({selected}.fx < -0.001) || ({selected}.fx > 1.001) ||
> ({selected}.fy < -0.001) || ({selected}.fy > 1.001) || ({selected}.fz <
> -0.001) || ({selected}.fz > 1.001))
>             { hide (hidden || {selected}); }
>     };
> delete hidden;
>
> Dean
>