Re: [Jmol-users] selecting atom IDs such as 1h47

[Timothy Driscoll]

On Oct 14, 2005, at 8:02 , Angel Herraez wrote:


Hello all
I have a PDB file in which some atoms have 4-character ID starting  
with a

number; Jmol does not allow me to select them
Example:
ATOM  1  N50 LIP 1 -22.659  21.125 -23.903   .00   .
ATOM  2  C47 LIP 1 -21.387  21.793 -24.221   .00   .
ATOM  3 1H47 LIP 1 -21.046  21.461 -25.101   .00   .
ATOM  4 2H47 LIP 1 -21.532  22.781 -24.272   .00   .
ATOM  5 3H47 LIP 1 -20.717  21.592 -23.507   .00   .

I can   select lip.c47
but not  select lip.1h47, I get this error: end of expression expected
If I try  select lip.?h47, it works (but of course I get selected  
more than I want)


Any clues? I have tested with 10.00.24  and 10.00.42


hi Angel,

just curious: that atom naming system is not familiar to me.  is it  
specific to lipids?


tim
--
Timothy Driscoll
molvisions - see. grasp. learn.

earth:usa:virginia:blacksburg






---
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
___
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users

Re: [Jmol-users] selecting atom IDs such as 1h47

[Miguel]

Angel & Miguel wrote:
>
>> Hello all
>> I have a PDB file in which some atoms have 4-character ID starting with
>> a number; Jmol does not allow me to select them
[snip]
>
> This looks like a fundamental problem with the Jmol atom expression
> parser.
>
> Having the integer first is causing problems.

I have taken a look at this. The problem is worse than I thought.

In the expression

select *.1h47

the '.1' is being recognized by the lexical parser as a floating point
number.

That means that this problem is more complicated than I expected. I do not
see an easy solution. I fear that it will require rethinking/rewriting the
expression parser. That will not happen in the near future.


Miguel

-
Open Source Molecular Visualization
www.jmol.org
[EMAIL PROTECTED]
-



---
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
___
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users

[Jmol-users] DOCUMENTATION: centerAt

[Miguel]

> On 17 Aug 2005 at 8:59, Miguel
> announced Jmol prerelease 10.00.32
>
>>  - default center of rotation is now the center of the bounding box
>> instead of the average point
>>  - centerAt provides additional flexibility for defining the center of
>> rotation
>
> This "centerAt" is not documented. Can anyone explain it?
> I am interested since I have had trouble with positioning some models.

centerAt absolute|boundbox|average {}

absolute specifies absolute angstroms coordinates
boundbox is relative to the center of the boundbox. The boundbox is
defined by the minimum and maximum atom center coordinates along each of
the cartesian axes.
average is relative to the average atom position. Also known as the
'unweighted center of gravity'

If coordinate values are not specified then they are assumed to be 0,0,0

# absolute cartesian coordinates 1.0, 2.0, 3.0
centerAt absolute 1.0 2.0 3.0;

# the following two are equivalent
centerAt absolute;
centerAt absolute 0 0 0;

# offset relative to the center of the boundbox
centerAt boundbox 10 20 -30

#the following are equivalent to the Chime + new default centering position
centerAt boundbox 0 0 0;
centerAt boundbox;
center;
center all;
center *;
select all; center selected;

# offset relative to the average atom position
# this was the default in Jmol 10.00
centerAt average 1 1 1;

# the following are equivalent
centerAt average 0 0 0;
centerAt average;


Miguel



---
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
___
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users

Re: [Jmol-users] pmesh questions

[Miguel]

>>Q: How would one decide where it would be displayed?
>>
> The default reference point could be the first coordinate in the pmesh
> file. Optionally the number of the reference point coordinates in the
> pmesh file could be specified in the pmesh definition line (e.g.: pmesh
> example1 "example1.pmesh" 17). The rest would then be similar to atom
> labels.

OK ... that is a pretty good idea.

>>Q: Could the labels be in one of the corners, in the same color as the
>>associated pmesh surface?
>>
>>
> If it would be in the corner in the same color it wouldn't be visible if
> the pmesh would be filled. And the label might also be too long to fit
> into the pmesh.

I meant in the corner of the Jmol display, not in the corner of the mesh.

>>I do not understand.
>>
> Maybe I should tell what we want to do. We want to highlight exon
> boundaries in protein structures. In order to be able to highlight for
> example PROSITE motifs at the same time the best way would be to show
> something between the two amino acids at the border.

I don't understand the details of this.

But I generally understand that this is some characteristic that you wish
to hilite.

> With pmesh this would be a filled square, perpendicular
> to the trace between the amino acids.

While that would certainly work, it does not sound like the kind of
application that pmesh was designed for.

> With pseudo atoms in the corners of the square and bonds from
> each atom to each other atom you would get a square with a cross inside.
> Not as good as a filled square but at least it could be labelled and all
> squares could be integrated into the PDB file as additional model.

OK

>>>Q: Is it possible to change the appearance of bonds besides changing
>>> their color?
>>
>>You can adjust color, radius, and translucency.
>>
>>hbonds are displayed as segmented (dashed) cylinders
>>
>>Q: What other renderings/formats would be interesting?
>>
>>
>>
> When I tested the pseudo atom pmesh replacement I just didn't realized
> the possibility to adjust the bond radius by using the wireframe command
> (although I used it plenty of times before!).
> For their regular use I guess cylinders are sufficient. For their
> 'misuse' a flat rectangle would be nice, but then you would have the
> orientation problem. So just forget about it.

It is not clear to me where you want to take this disussion.


Miguel



---
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
___
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users

Re: [Jmol-users] pmesh questions

[Timothy Driscoll]

On Oct 19, 2005, at 12:37 , Rolf Huehne wrote:



Maybe I should tell what we want to do. We want to highlight exon
boundaries in protein structures. In order to be able to highlight for
example PROSITE motifs at the same time the best way would be to show
something between the two amino acids at the border. With pmesh this
would be a filled square, perpendicular to the trace between the amino
acids. With pseudo atoms in the corners of the square and bonds from
each atom to each other atom you would get a square with a cross  
inside.
Not as good as a filled square but at least it could be labelled  
and all

squares could be integrated into the PDB file as additional model.



this sounds pretty cool.  so you want something like this:

AA1---AA2--AA3-|-AA4--AA5...

where the pipe represents your motif border?

it seems to me that it might be simpler to add a new kind of  
annotation object (like a label, but a filled rectangular cube).  i.e.,


set blocksize 12
color blocks red
block aa3,aa4


any thoughts?

tim
--
Timothy Driscoll
molvisions - see. grasp. learn.

earth:usa:virginia:blacksburg






---
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
___
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users

Re: [Jmol-users] pmesh questions

[Rolf Huehne]

Miguel wrote:
>>Hi all,
>>
>>I tried the "pmesh" functions of Jmol (10.00.36) and have a few questions.
>>
>>Currently there seems to be no labelling possibility for pmesh objects.
>>
>>Q: Would you consider adding a pmesh labelling command?
>>
>
>It is not clear to me what kind of label you want.
>
>Q: Do you want a label that would show up on the display?
>  
The same kind of label as for atoms, permanently shown (Are there other
kinds? Or do you mean 'hover'?).
>Q: How would one decide where it would be displayed?
>  
The default reference point could be the first coordinate in the pmesh
file. Optionally the number of the reference point coordinates in the
pmesh file could be specified in the pmesh definition line (e.g.: pmesh
example1 "example1.pmesh" 17). The rest would then be similar to atom
labels.
>Q: Could the labels be in one of the corners, in the same color as the
>associated pmesh surface?
>
>  
If it would be in the corner in the same color it wouldn't be visible if
the pmesh would be filled. And the label might also be too long to fit
into the pmesh.
>>In order to address pmesh objects individually it is necessary
>>to put them into separate files. This is not very efficient
>>if you have dozens of objects
>>and need to contact the server for each object.
>>
>>Q: Would you consider adding support for a multiple pmesh format to be
>>able to load all pmesh objects at once?
>>
>
>Probably not.
>
>The pmesh format is a standard format. I do not want to create new file
>formats.
>
>I don't think there is much inefficiency introduced by having to fetch
>multiple files from the file server. If the .pmesh files are small then it
>is not a big deal. If the .pmesh files are large then it really does not
>matter.
>
>Web servers are quite efficient. Go to any commercial web site and dozens
>of image files are downloaded, each of which is downloaded separately from
>the web server.
>
>  
And then you will notice the difference if you switch off image
download, depending on the usage and settings of the server.
But the web server part is only one end of the story. Since we are
providing information for all ~34000 PDB entries it would make a
considerable difference to have additionally 34000 other files or
34. But there are of course always several  roads to rome...

>>The minor alternative to pmesh would be to add pseudo atoms to the
>>coordinate
>>file (which could be labelled) and use bonds to build up some kind of
>>objects.
>>
>
>I do not understand.
>  
Maybe I should tell what we want to do. We want to highlight exon
boundaries in protein structures. In order to be able to highlight for
example PROSITE motifs at the same time the best way would be to show
something between the two amino acids at the border. With pmesh this
would be a filled square, perpendicular to the trace between the amino
acids. With pseudo atoms in the corners of the square and bonds from
each atom to each other atom you would get a square with a cross inside.
Not as good as a filled square but at least it could be labelled and all
squares could be integrated into the PDB file as additional model.
>  
>>Q: Is it possible to change the appearance of bonds besides changing their
>>color?
>>
>
>You can adjust color, radius, and translucency.
>
>hbonds are displayed as segmented (dashed) cylinders
>  
>Q: What other renderings/formats would be interesting?
>
>
>  
When I tested the pseudo atom pmesh replacement I just didn't realized
the possibility to adjust the bond radius by using the wireframe command
(although I used it plenty of times before!).
For their regular use I guess cylinders are sufficient. For their
'misuse' a flat rectangle would be nice, but then you would have the
orientation problem. So just forget about it.

Regards,
Rolf


---
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
___
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users

Re: [Jmol-users] PFAAT uses Jmol

[Timothy Driscoll]

On Oct 19, 2005, at 11:01 , [EMAIL PROTECTED] wrote:


De: Timothy Driscoll <[EMAIL PROTECTED]>


hi all,

in case anyone is interested:

I just came across a protein sequence alignment tool called PFAAT
().  this software uses Jmol as an
inline structure visualization module.  it looks pretty cool, and I
can see great possibilities in the combination. (displaying aligned
3D structures, coloring 3D structure by alignment score, etc.)



Thanks for the link :)
You should add it in the Jmol wiki at http://wiki.jmol.org/ 
ApplicationsEmbeddingJmol




done. :-)

tim
--
Timothy Driscoll
molvisions - see. grasp. learn.

earth:usa:virginia:blacksburg






---
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
___
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users

Re: [Jmol-users] PFAAT uses Jmol

[nvervell]

De: Timothy Driscoll <[EMAIL PROTECTED]>
>
>hi all,
>
>in case anyone is interested:
>
>I just came across a protein sequence alignment tool called PFAAT  
>().  this software uses Jmol as an  
>inline structure visualization module.  it looks pretty cool, and I  
>can see great possibilities in the combination. (displaying aligned  
>3D structures, coloring 3D structure by alignment score, etc.)

Thanks for the link :)
You should add it in the Jmol wiki at 
http://wiki.jmol.org/ApplicationsEmbeddingJmol

Nico



---
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
___
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users

Re: [Jmol-users] PFAAT uses Jmol

[Miguel]

> hi all,
>
> in case anyone is interested:
>
> I just came across a protein sequence alignment tool called PFAAT
> ().  this software uses Jmol as an
> inline structure visualization module.  it looks pretty cool, and I
> can see great possibilities in the combination. (displaying aligned
> 3D structures, coloring 3D structure by alignment score, etc.)

I have had contact with the PFAAT developers.

They are planning to contribute some modifications to the Jmol source in
order to make things work better with PFAAT.


Miguel



---
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
___
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users

[Jmol-users] PFAAT uses Jmol

[Timothy Driscoll]

hi all,

in case anyone is interested:

I just came across a protein sequence alignment tool called PFAAT  
().  this software uses Jmol as an  
inline structure visualization module.  it looks pretty cool, and I  
can see great possibilities in the combination. (displaying aligned  
3D structures, coloring 3D structure by alignment score, etc.)


bye,

tim
--
Timothy Driscoll
molvisions - see. grasp. learn.

earth:usa:virginia:blacksburg






---
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
___
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users

Re: [Jmol-users] Update Request for 'Monitor / Measure' command

[Miguel]

> In the category of wish lists, may I propose the following?
>
> Currently "monitor" ("measure") uses AtomNo for each of its parameters.
>
> As I develop a single page to service an Applet that
> can dynamically load different, related structures,
> the monitor command, of course, is less useful.

Yes. I understand the problem. This is an issue that has bothered me for
some time.

> The S on Cys35, chain B has different atom numbers
> in the two structures. Most of my other commands work fine in
> this context.
>
> Might I suggest that the Monitor/Measure algorithm optionally take a
> unique atom descriptor for each of its arguments?
>
> Example: monitor 184:A.CA 367:A.CA

Your *protein* bias is showing ... we need to have a format that would
work for all file types ;-)

Here are the thoughts that I have had on this subject in the past:

'measure' should take a set of atoms, where the set is defined using the
normal 'atom expression' syntax. Unfortunately, this will only work for
distance measurements. Angles and torsions must be ordered, and sets are
unordered collections of atoms.

The first solution I tried to come up with was to try and maintain an
order within small sets. That is, if the set size is small then remember
the order in which the atoms were selected. The boolean set operators
(and, or, not) would make a 'best efforts' attempt to maintain some type
of order that made sense.

However, I think I concluded that this would be error-prone and would
break down as people tried to work with more complicated atom expressions.

The other solutions I came up with was:

Allow multiple sets to be passed to the 'measure' command. The scripting
language grammer is rather poorly defined. To enable accurate parsing each
set would have to be fully enclosed in parentheses. (note that the
polyhedra command takes an optional parenthesized atom set)

In the general case, the atoms sets would be required to have exactly one
atom each. Using your example from above:

  measure (184:A.CA) (367:A.CA);

In the special case of two atoms, we might choose to allow the following:

  monitor (184:A.CA,367:A.CA);

which could also be written as:

  select 184:A.CA,367:A.CA;
  monitor selected;

Having the extra parentheses seems somewhat verbose, but given the grammar
that we started with, I don't see much other choice.


Tell me what you think.


Miguel




-
Open Source Molecular Visualization
www.jmol.org
[EMAIL PROTECTED]
-



---
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
___
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users

Re: [Jmol-users] pmesh questions

[Miguel]

> Hi all,
>
> I tried the "pmesh" functions of Jmol (10.00.36) and have a few questions.
>
> Currently there seems to be no labelling possibility for pmesh objects.
>
> Q: Would you consider adding a pmesh labelling command?

It is not clear to me what kind of label you want.

Q: Do you want a label that would show up on the display?

Q: How would one decide where it would be displayed?

Q: Could the labels be in one of the corners, in the same color as the
associated pmesh surface?

> In order to address pmesh objects individually it is necessary
> to put them into separate files. This is not very efficient
> if you have dozens of objects
> and need to contact the server for each object.
>
> Q: Would you consider adding support for a multiple pmesh format to be
> able to load all pmesh objects at once?

Probably not.

The pmesh format is a standard format. I do not want to create new file
formats.

I don't think there is much inefficiency introduced by having to fetch
multiple files from the file server. If the .pmesh files are small then it
is not a big deal. If the .pmesh files are large then it really does not
matter.

Web servers are quite efficient. Go to any commercial web site and dozens
of image files are downloaded, each of which is downloaded separately from
the web server.

> The minor alternative to pmesh would be to add pseudo atoms to the
> coordinate
> file (which could be labelled) and use bonds to build up some kind of
> objects.

I do not understand.

> Q: Is it possible to change the appearance of bonds besides changing their
> color?

You can adjust color, radius, and translucency.

hbonds are displayed as segmented (dashed) cylinders

Q: What other renderings/formats would be interesting?


Miguel



---
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
___
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users

[Jmol-users] Update Request for "Monitor / Measure" command

[Craig T Martin]

In the category of wish lists, may I propose the following?

Currently "monitor" ("measure") uses AtomNo for each of its parameters.

As I develop a single page to service an Applet that can dynamically  
load different, related structures,
the monitor command, of course, is less useful.   The S on  
Cys35, chain B has different atom numbers
in the two structures. Most of my other commands work fine in  
this context.


Might I suggest that the Monitor/Measure algorithm optionally take a  
unique atom descriptor for each of its arguments?


Example: monitor 184:A.CA 367:A.CA

Thanks,

Craig Martin
[EMAIL PROTECTED]





---
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
___
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users

[Jmol-users] pmesh questions

[rhuehne]

Hi all,

I tried the "pmesh" functions of Jmol (10.00.36) and have a few questions.

Currently there seems to be no labelling possibility for pmesh objects.

Q: Would you consider adding a pmesh labelling command?

In order to address pmesh objects individually it is necessary to put them into
separate files. This is not very efficient if you have dozens of objects and
need to contact the server for each object.

Q: Would you consider adding support for a multiple pmesh format to be able to
load all pmesh objects at once?


The minor alternative to pmesh would be to add pseudo atoms to the coordinate
file (which could be labelled) and use bonds to build up some kind of objects.

Q: Is it possible to change the appearance of bonds besides changing their
color?

Regards,
Rolf



This message was sent using IMP, the Internet Messaging Program.


---
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
___
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users

Re: [Jmol-users] default zoom has changed along versions

[Angel Herraez]

El 16 Oct 2005 a las 19:45, Miguel escribió:
> This is an intentional change.

Great!  I think this change is better.
I will have to revise the versions in my website.
· · · · · · · · · · · · · · · · · · · · · · · · · · · · ·
 Dr. Angel Herraez
 Dep. Bioquimica y Biologia Molecular, Universidad de Alcala
 E-28871 Alcala de Henares  (Madrid), Spain
 fax: +34-91 885 45 85



---
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
___
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users