Re: [Jmol-users] Bug? in select by atom number?

[Rolf H. <rh...@fl...>]

On 06/11/2012 06:21 PM, Gutow, Jonathan H wrote:
> I am trying to explain how the select statement in 12.2.28, works to one of my students and am confused by the following example.
>
> load =1lcd
> select ({553:559})
>
> This selects the carbons in a phenyl ring plus the C it is attached to.   Here's what has me confused.  The atoms selected are:
>
> #556, #557,#558,#559, #560, #561, #562
>
> These are clearly not within the selection range indicated.  My first interpretation is that it maybe this is based on connectivity, because it selects atoms connected to #553 following bonds to #559.  But that's not what it is doing either.  If you try:
>
> select ({557:559})
>
> You get three atoms selected #560, 561 and 562.  It certainly is three atoms, but why these three and not #558, 560 and 561, which are nearest neighbors in the ring to #557 rather than 559?
>
> How is Jmol interpreting these select statements?  I'd like to be able to explain this too him.
>
The syntax you used "select ({553:559})" uses the unique atom index 
numbers assigned by Jmol to each atom. These are different from the atom 
numbers provided in a PDB file.

You would need a command like "select atomno>=553 and atomno<=559" to 
use the "official" atom numbers.

You can see the difference by labelling the atoms with both numbers:
   label "%a ind=%D atomno=%i"

Regards,
Rolf