Rolf, first, these are just development versions (11.5.x), so we can
undo anything. I did see your suggestion, but I don't think it is really
workable. You can't have an algorithm that sometimes does one thing and
sometimes another in such an unpredictable fashion. (And, remember, this
is chain-specific.)
The logic here is that (a) we can have it the way we always did -- using
that flag, and (b) we can have it the way a user would expect as well.
That's the general solution to this. So the real question is whether we
want the default for sequenceRangePhysical to be TRUE or FALSE.
I'm hearing you say you want it TRUE, so that there are no changes to
current scripts. That's certainly the way I usually lean, but in this
case I thought Eric made a significant argument for changing the default
behavior to one that is far more logical to the user. (The user
generally doesn't care what the physical layout of the file is -- they
just want to know about residue numbers, and since the proposed change
is consistent with Rasmol behavior, one could argue that what we have
now is a bug, in fact.)
As for Insertion codes....
Rolf Huehne wrote:
>
>>
>>
>
>Maybe I havn't understood the insertion code part correctly, but for me
>it looks like there will be no range selected in the above example
>"1-3^A". Suppose the sequence would be "1,2,3,3A". Then the only residue
>selected would be "3A" because it is the only one that matches all three
>conditions.
>
>
>
Basically, I propose that what "1-3^A" means is "within range 1-3, all
inserted residues of type A". That's easy to explain. Certainly in some
cases there could be a string of these, so it could amount to a range.
So, for example, with 2by9, we get (old/your) way:
load =2by9
select 182-183
$ select 182-183 and *.N
$ print {selected}.label()
[CYS]182:X.N #1269
[ALA]183:X.N #1275
but 183-184:
$ select 183-184
$ print {selected}.label()
[ALA]183:X.N #1275
[GLY]183^A:X.N #1280
[TYR]184:X.N #1284
That seems strange and rather unexpected to me. It's because the
inserted item is after the original in the file. I would rather see (I
think):
$ select 182-183 and *.N
$ print {selected}.label()
[CYS]182:X.N #1269
[ALA]183:X.N #1275
[GLY]183^A:X.N #1280
$ select 183-184
$ print {selected}.label()
[ALA]183:X.N #1275
[GLY]183^A:X.N #1280
[TYR]184:X.N #1284
(Actually, I have some question as to whether the alternative locations
should be included. I think so.)
Eric, if you "select 183" do you expect to select it and the inserted
code next to it? How about 183-184 -- what would you expect there?
183, 183^A, 184, 184^A
or
183, 183^A, 184
or
183, 184
?
>Bob, you havn't commented yet on my last proposal (use "<= / >="
>notation only if one of the border residues is missing). I still think
>this is the best solution that was mentioned yet. Of course the physical
>range selection is the only way to cope reliably with numbering
>irregularities like insertion codes. So any "<= / >=" based method will
>fail at least in some cases.
>
>
Don't worry about my committing it or not -- let's arrive at a consensus
view if we can. I have to say I don't like the "minimize the failure
rate" idea. Let's get this fixed.
>And this is also true for the solution of Jmol 11.5.30 (if I understood
>it correctly). But it has 2 additional disadvantages:
>
>1) some people have to change scripts/sites to revert to a correct selection
>
>
>
I'm not convinced that will be a problem. I just don't see how we can
have it behave one way and then another simply based on knowledge the
user doesn't have (whether groups are present or not in a specific chain).
>2) one has to decide principally which method should be used
>
>
>
If it's a serious concern, all we need to do is have the default
behavior be
set sequenceRangePhysical TRUE
and let Eric be sure to set sequenceRangePhysical FALSE for his pages.
>My proposal lacks both of these disadvantages and still reduces the
>selection failure rate.
>
>
>
my goal is to eliminate the failure rate.
Bob
>Regards,
>Rolf
>
>
>-------------------------------------------------------------------------
>This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
>Don't miss this year's exciting event. There's still time to save $100.
>Use priority code J8TL2D2.
>http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
>_______________________________________________
>Jmol-users mailing list
>[email protected]
>https://lists.sourceforge.net/lists/listinfo/jmol-users
>
>
--
Robert M. Hanson
Professor of Chemistry
St. Olaf College
Northfield, MN
http://www.stolaf.edu/people/hansonr
If nature does not answer first what we want,
it is better to take what answer we get.
-- Josiah Willard Gibbs, Lecture XXX, Monday, February 5, 1900
-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
Jmol-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/jmol-users
