Re: [Jmol-users] Selection problem with structure 1TON

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

Miguel wrote:
> Rolf wrote:
>   
>>> As an example, let us assume that we have a chain with the following 10
>>> residue sequence numbers.
>>>
>>> 1 2 3 4 5 1 2 3 4 5
>>>
>>> Q: If I say 'select 1-5' should all 10 residues be selected?
>>>
>>>       
>> Generally I would expect all 10 residues to get selected.
>>     
>
> I think that you are thinking all 10 residues for a different reason than
> I was thinking.
>
> I was thinking that there were two ranges from 1-5.
>   
I agree, non greedy but global
> You were thinking of one range that ran fromm the first 1 to the last 5.
>
>   
no, because this case must be two different chains with a TER record in 
between.
>   
>>  > Q: If I say 'select 2-3' should 4 residues be selected?
>>  >
>> This is even more complicated, because you have 3 alternatives:
>>    2 or 4 or 7 residues
>>     
>
> Correct.
>
> Jmol today will return 2 in this case because the implementation assumes
> that residue numbers are unique.
>   
only the atom number is unique in a PDB file, therefore again we have 
two sets (from two chains, or hetero) with two members each, 4 residues 
should be selected.
>   
>> It raises the additional question if the range should be considered as
>> local (2,3 and 2,3) or as global (2,3,4,5,1,2,3).
>>
>> As a default I would expect it to be considered as local and greedy.
>>     
>
>   
I disagree! I vote for ((2,3) , (2,3)) and regard this as non greedy but 
global (all occurrences)
ignoring the chain structure in the PDB file would be a very bad idea.
Please Miguel, don't do that.
With kind regards, Jan
> OK
>
> Well that is the answer that will give you the behavior you want with the
> file that caused all these problems.
>
> I will see what I can do.
>
>
> Miguel
>