[Jmol-developers] Isosurfaces in the context of multiple files

[Bob H. <ha...@st...>]

Now, wait a minute.... I have just checked this in.... It should not be 
working yet for you, Tim.... It's late; you must be half asleep. I'm not 
THAT fast. :)

Isosurfaces in the context of multiple models/files
-----------------------------------------

When multiple models and or multiple files are loaded and then 
isosurfaces created, there can be some problems. We should keep alert to 
bugs in this area, as it is one of them most complicated parts of Jmol. 
The way it SHOULD (and I think does in the next prerelease) work is as 
follows:

Isosurfaces are generally associated with specific models. A solvent 
surface, a molecular orbital, etc. There are exceptions, but let's not 
worry about those yet.

Provided you have a single model displayed, as with

model 3.1

or

frame 3

then any isosurface creation that is done should automatically restrict 
itself to that model. You don't have to select the model, you don't have 
to "ignore" atoms that are not in the model. Jmol should automatically 
does that for you. There was a bug there I just fixed. The following 
sequence should do just what you think it should do:

load files "=1crn" "=1blu" "=1dry"
model 1.1;isosurface s1 select(protein) solvent 0.8
model 2.1;isosurface s2 select(protein) solvent 0.8
model 3.1;isosurface s3 select(*) solvent 0.8

This creates three isosurfaces, one in each frame 1.1, 2.1, and 3.1.

If you don't want to switch frames, you can put the model designation 
right into the isosurface command:

load files "=1crn" "=1blu" "=1dry"
isosurface s1 model 1.1 select(protein) solvent 0.8
isosurface s2 model 2.1 select(protein) solvent 0.8
isosurface s3 model 3.1 select(*) solvent 0.8

This should create the same set of isosurfaces as above. The only 
difference is that in the first example you are left looking at model 
3.1 and isosurface s3; in the second example you are left looking at the 
frame you were looking at before the commands were given -- with 
whatever isosurface might be created in that frame.

Of course, if you use

frame 0

then you see all three isosurfaces. Now, what do you suppose happens 
with this next example?

load files "=1crn" "=1blu" "=1dry"
frame 0;
isosurface s1 select(protein) solvent 0.8

We are displaying all three models; we are creating a single isosurface. 
Now what? Well, something has to give. Jmol just displays the isosurface 
for the first model, 1crn.

"fixed" isosurfaces
------------------

What if we want to create an isosurface for one model but for whatever 
reason display it not with its atoms but with those of another model? 
This is a bit ghostly, but if you add the keyword FIXED to the 
isosurface command:

load files "=1crn" "=1blu" "=1dry"
frame 2.1;
isosurface s1 fixed select(protein) solvent 0.8

what happens is that the isosurface for 1blu is created as expected, but 
it is fixed in space for all models, not just the one. So you can switch 
from one frame to the next and that isosurface will be there. It's a bit 
odd, I know. but it's particularly useful with something simple, like a 
plane:

load files "=1crn" "=1blu" "=1dry"
isosurface s1 fixed plane x=13

This is nice, because it provides a reference plane for all models.

Bob