Quoting Miguel <[EMAIL PROTECTED]>:
Quoting Miguel <[EMAIL PROTECTED]>:
It is not clear to me where you want to take this disussion.
I started this discussion to help us to decide what we should
use for highlighting boundaries between two amino
acids: 'pmesh' or 'pseudo atoms + bonds'.
Please be patient with me.
Remember that I am not a chemist and do not understand anything about the
applications that you are trying to build.
I have no idea what the 'boundary between two amino acids' is.
The boundary is nothing real in a biochemical way. It is just an abstraction
used to mark the end of one region and the beginning of another region.
Like an
invisible fence between two houses: house 1 is amino acid 1 and house 2
is amino
acid 2. And the pmesh would be the fence made visible.
All I understood was 'can I put a label on a pmesh'?
With the current state of Jmol we could only use pmesh
combined with a pseudo atom for labelling. Therefore we
would most probably use 'pseudo atoms + bonds' although the
highlighting wouldn't be as good as with 'pmesh'.
I am getting confused.
Because I don't want to force adding features for 'misusing'
bonds I didn't went further in that direction.
Good. I do not like to misuse features.
If pmesh labelling would be added we would presumably
use 'pmesh', even with a lot of different files.
But this sounds like you would be misusing the pmesh command
I don't have any idea what the developers of the pmesh format had in
mind, when
they developed it. From the discussions on the Jmol mailing lists I got the
impression that it was intended for building surfaces. At least the example
shown on the "Jmol interactive scripting documentation" page
(http://www.stolaf.edu/people/hansonr/jmol/docs/#pmesh) doesn't seem to be
qualitatively different from what we wanted to use it for.
I like Timothy's idea of a new annotation object best, because
it would make any precomputing and/or changing of the PDB files
unnecessary.
Tim wrote:
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
But I guess it would be more work than adding a lablling
command for 'pmesh'.
I am in favor of adding features if they are generally useful.
We need to restart this discussion from the beginning.
Draw me a picture of what you want. Use a graphics program or use a pencil
& paper and scan the results.
I prepared an example Jmol page that uses 'pmesh' or 'pseudo-atoms+bonds' to
highlight the boundary between amino acid residue 8 (red, chain A) and amino
acid residue 9 (blue, chain A). You can switch either highlighting type on/off
by using the buttons at the top of the example page:
http://www.fli-leibniz.de/ImgLibPDB/jmol/pmesh_example1.html
The coordinates of the corners of the rectangle are not optimized yet. I think
ideally it should be a square that is perpendicular to the trace drawn between
the to amino acid residues (representing the backbone of the protein). We
havn't achieved this yet.
In the example we used the alpha carbon atoms of the two amino acid
residues as
reference points. The point in the middle of a straight line between these
points is the centre of the rectangle. The beta carbon atom of one amino acid
residue was then used as a third reference point to calculate the coordinates
of the corners.
Tim sent me this url:
http://www.expasy.org/cgi-bin/prosite/PSView.cgi?ac=PS50020&onebyarch=1&hscale=0.6
But I do not understand how this relates to your request.
I think this should help you to understand what "PROSITE motifs" are (which I
named as something we want to be able to highlight simultaneously with the
amino acid boundaries). But I guess the page requires to much
biological/biochemical knowledge to help you.
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
[email protected]
https://lists.sourceforge.net/lists/listinfo/jmol-users
