Re: [Jmol-users] connect and wireframe

[Jan <jo...@gm...>]

Bob Hanson wrote:
> I thought I should start a new thread for this. I don't mean to rush 
> anybody. Whatever timeframe works is fine by me.
>
> 1. Please check in if you have a stake in this.
> 2. Respond to proposals and/or propose your own.
I would prefer to stay close with the RasMol syntax:
RasMol>bound 512 513 +
Jmol>bound <src atomnumber|(src atom selection)> <target
atomnumber|(target atom selection)> +
# would increment bound order
Jmol>bound <src atomnumber|(src atom selection)> <target
atomnumber|(target atom selection)>
[SINGLE|DOUBLE|TRIPLE|AROMATIC|HBOND|DELETE]
# would set the specified bound order

> 3. We'll discuss them.
as two selection expressions in one command are currently not supported
in Jmol, I'm pleased with the current:
>
> OPTION 00.48 (current):
>
> select (atom expression)
> connect <max distance> (target atom expression)
> connect <bond type>
> connect DELETE
> where <bond type> includes SINGLE, DOUBLE, TRIPLE, AROMATIC, HBOND
with the small modification of an additional <bond type> INCREMENT.
INCREMENT should be the default.
With this default, it would be easy to transform any RasMol script in a
single path.

>
> properties:
> .Bonds that are already made are unaffected.
Bonds that are already made are incremented, regardless whether
meaningful or not
> .Bonds are introduced as SINGLE bonds only.
new bonds are introduced as SINGLE bonds
> ."connect <bond type>" and "connect DELETE" are dependent   upon 
> setting of BONDMODE:
>     set bondmode AND (connect WITHIN the selected set only)
>     set bondmode OR (connect WITHIN and TO the selected set)
This is tricky,
it would be much easier, if the previous incremented bonds are the 
selected bonds until a new selection is done (see below) (if necessary - 
no bond selection possible, only atom selection - with set bondmode AND 
a connect command would alter the selection to include all touched 
bonds, but what to do if bondmode is OR? a side effect of connect would 
be to alter bondmode automatically to AND?)

an additional <min distance> is not forcing
select carbon
connect 1.7 (oxygen)
connect SINGLE
select carbon
connect 1.5 (oxygen)
connect DOUBLE
>
> -----------------------------------
> OPTION BH.1:
>
> set connectmode <bond type>
> select (atom expression)
> connect <min distance> <max distance> (target atom expression)
>
> where <bond type> includes SINGLE, DOUBLE, TRIPLE, AROMATIC, HBOND, NONE
NONE is a bad choice for DELETE or UNBOUND (may be, none should restore
the previous state, after the file was read)
>
> properties:
> .Bonds that are already made are revised.
> .Bonds are introduced exactly as desired, not just as single bonds.
> .Actions are independent of the setting of BONDMODE
>
> Rationale:
> 1. You can be as specific as you want in selecting the atom sets and 
> now the distance range, so that suggests to me that there should be no 
> surprises. The command does just what it says and not 
> "connect-unless-not-already-connected."
The same is true for INCREMENT followed by a more specific
connect <bond type>
command
> 2. It needs no dependency upon BONDMODE because the distance ranges -- 
> that typically really do define bond order -- are the determining 
> factors.
successively defining higher bond orders would do the same thing
> 3. It is also more flexible than using BONDMODE, because now
> we have a specific starting set and a target set between which
> bonds will be built or removed.
the default
set bondmode and
and as side effect of connect would do the same
> 4. BONDMODE is a carryover from Rasmol, specifically for use with
> WIREFRAME and COLOR so as to select specific bonds. It's a rarely
to select bonds by selecting atoms.
> used command that isn't necessary if you can specify target sets. So, 
> for example, "set bondmode AND" is equivalent to:
>
>  select (some atoms)
>  connect ... (selected)
>
> And "set bondmode OR" is equivalent to
>
>  select (some atoms)
>  connect ... (all)
>
> Thus, connecting atoms doesn't need to depend upon BONDMODE. I suggest 
> that disconnecting them and changing their bond type
> should not as well.
I agree, after a connect command the connected bonds should be the
selected bonds until a new selection is made, that would be much more
straight forward.
Regards, Jan
>
>
> examples:
>
> select (carbon)
> set connectmode DOUBLE
> connect 1.3 1.5 (oxygen)
> set connectmode SINGLE
> connect 1.5 1.7 (oxygen)
> set connectmode NONE
> connect 1.7 1.9 (oxygen)
>
>
>
> Bob