Miguel wrote:
Jan wrote:
I would prefer to define a wild card for exact one alpha or number, this
offers more flexibility in other selection context, too.
I think that allowing single char wild card for this field could be
valuable, but I do not think that it solves the problem.
currently we have:
* the whole field
? one position (but could be empty)
and /0 in case of no MODEL is defined
Jmol defines the default model to be /1, not /0
should this be altered in Jmol for compatibility?
e.g. **:*.????/0 should be a correct RasMolScript expression
In Jmol this is **:*.????/1
with _ as existing alpha and number wild card (should it be
/[a-zA-Z0-9]/ or /[a-zA-Z]|([0-9]+)/ ?)
select not *;_
will select all atoms which have no altLoc
select */_
(my mistake)
will select all defined models (which isn't of great use)
I see no reason to use _ as a single char wild card. We already have ?
which we can use as a single char wild card in the altLoc context.
? could be empty and is of no use in this case, only *;_ will select all
defined altLoc atoms separately
and could shorten
select *.N? and not *.N
to
select *.N_
OK, this isn't really forcing
to complete the analogy with the MODEL it may be desirable to address
the whole "alternate location model" e.g.
*%A
which is in RasMolScript notation (if there were only *;A and *;B
alternate conformations) equivalent with
select not *;B
Yes.
I am thinking that there is a need to address the 'altLoc' atoms as well
as the entire model.
select *%A ; selects only the altloc=='A' atoms
OK, but as long there is no decision I'll use RasMolScript notation
select *;A
select */A ; selects the entire 'A' model ... altloc==' ' || altLoc=='A'
OK,
in this case, because MODEL is an integer, but what about if altLoc is a
digit,
I was going to remove the limitation that MODEL be an integer.
what about the possible cases where model and altLoc coexists?
In 1SKT we have e.g.
select [LEU]14.HD2;1/1
Good point.
I propose the following:
if no altLoc is present and MODEL is present then the model names will be
the MODEL numbers.
if altLoc is present and no MODEL is present then the model names will be
the altLoc names.
if both altLoc and MODEL are present then the model names will be
<altLoc>_<MODEL>
why wasting an extra identifier, I would argue to use the same (you may
say %, but for now, I'll stay with ;) char as model name separator again
and exchange positions:
/<MODEL>;<altLoc>
(or if you insist /<MODEL>%<altLoc>) to address the entire model
So, if a file has altLocs 'A' and 'B' in MODELs 1 and 2 then the 4 model
names within Jmol will be
A_1
A_2
B_1
B_2
To select the 'A' altLoc atoms you can use
select *%A
select *;A
(only the position before or after / will define the meaning altLoc
atoms only or the entire altLoc model:-)
To select the entire model in MODEL 2 you can use
select */A_2
select */2;A
which will (for compatibility) be the same as (with the first default
altLoc)
select */2
in contrast to
select */2;B
which is not default
model A_2
model=2;A
Regards, Jan
-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games. How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Jmol-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/jmol-users
