Re: [Jmol-developers] new bond command

[Miguel <mi...@jm...>]

Jan wrote:
> The RasMol script command to create bonds is called bond.
> bond <source> <dest> +
> unbond <source> <dest>
...
> bond because it is the RasMol script command

Since RasMol uses the command 'bond' then I would feel some obligation to=

use the same syntax & semantics.

As I said, I do not like using atom numbers to specify bonds.


>> The 'link' command should use atom sets to specify the atoms.
>>
> as the monitor command :-)

Agreed.


>> Q: Thoughts?
>>
> why not
> bond <src atom selection> <dest atom selection> =5B+-=5D
> e.g.
> bond (=5BCYS=5D1.CA) (=5BARG=5D2.CA)
> bond (carbon) (hydrogen and within(1.5, carbon))

The evaluation engine generally assumes that there is a 'current' set in
the accumulator.

In addition, there are no script commands that take more than one atom
expression set as an argument.

Therefore, it would be difficult to make it work in this manner.


> yes, distance is important and an iterating distance command would be
> superior over the within command.
> let distant evaluate <source> and <dest>
> bond distant(1.5, (carbon) (hydrogen))

I do not see this as an improvement over:

  select carbon; link hydrogen and within(1.5, carbon);

> I would say, the command without distance is useless, distance need not=

> to be optional.

I am now thinking that the distance is not necessary. Everything can be
done using the existing atom expression operators.


>>   wireframe single =7C double =7C triple =7C aromatic =7C hydrogen =7C=
 <diameter>
>>
> With RasMol script, I use the bond command before the wireframe command=
,
> if I want to draw a ligand with correct bond order.

I do not like the RasMol script implementation of

  bond <atom-number-1> <atom-number-2> +

where it increases the bond order if the bond already existed.


Miguel