Hi Michael,
Please see my comments below...
Hi Séb
Thank you very much for your suggestions. See my comments below.
Cheers
Michael
Sébastien Morin schrieb:
Hi Michael,
Here are a few details I thought of...
1.
I guess you should try to stick to the nomenclature used in relax. In
particular, T1 and T2 could be changed to R1 and R2.
I will change these labels according to 'relax'.
2.
How are R1 and R2 duplicates handled ? What about duplicated NOE spectra
for error estimation instead of noise RMSD ?
relaxGUI compares the values of entered relaxation times and sets the
dublicates. This is done by looping over the relaxation times and
comparing them to each other. For the NOE, I adapted the procedure as
suggested by the relax manual.
Oups... I was almost sure the duplicate NOE data to extract errors was
already implemented... I can't remember exactly, but I think we
discussed that some time ago... Anyway, your implementation is as in
relax right now...
3.
A sequence could be entered in place of a pdb file for NOE, R1 and R2
calculations. This could be useful for people working on a protein of
unknown structure. Nevertheless, the structure could also be useful for
displaying the results with some color scheme.
There is a need for the structure for model-free analysis, therefore I
think it is also usefull, if the noe values are colour-coded, as in the
upload of today. But I will think about entering the sequence (don't
know how to feed it into relax, yet).
There is a function called 'sequence.read'...
4.
What internal script (or approach) does the GUI use for model-free
calculations ? Is it that from the 'full_analysis.py' script ?
it is basically the full_analysis.py script, just with more variables.
5.
Is there a section for setting diffusion tensor details ?
what are you thinking about? might be interesting!
Well... If you use the 'full_analysis.py' script, then different
diffusion tensors will be optimised, depending on the progress of the
analysis.
Right now, the script will run until convergence until a defined
diffusion tensor is completely optimised, starting with the 'local_tm'
diffusion tensor.
Hence, the GUI could verify if results for different tensors are present
and, then, allow the analysis to go on. For example, without any
results, the only available analysis would be that of 'local_tm'. After,
the four diffusion tensors 'sphere', 'prolate', 'oblate' and 'ellipsoid'
would be available. Once at least one is completed, the final run for
optimisation of all parameters at once (and Monte Carlo simulations for
error estimation) would be allowed...
What do you think ?
For a new user, I guess this kind of approach would make sens and allow
a better understanding of the analytical scheme...
The rest is fine for now !
Cheers,
Séb :)
6.
AICc model selection could be added in the model-free section.
7.
In the global settings windows, there could be proposed values for the
minimisation algorithm.
will be added
8.
Does the button 'Contact relaxGUI' work ?
not yet
9.
The button 'Reference' could be written with an 's'... :p
not yet. t
10.
Does the button 'Manual' work ?
not yet. the settings menu is also not implemented
This is all I see right now...
This GUI is an impressive work, Michael !
Cheers !
Séb :)
Michael Bieri wrote:
Hi
wx.Python is the module of all graphical objects, such as text boxes,
buttons, panels... I thought wx.python is installed as standard in
linux
distributions (I think it is on Ubuntu). Therefore, I used these
modules
to build the GUI.
I will be on holiday from tomorrow until next Wednesday, so I will
work
on the incorporation next week. If you have some ideas what kind of
parameters should be included in the Extras > Settings... dialog,
please
let me know and I will add them.
Cheers
Michael
Edward d'Auvergne schrieb:
Hi,
I received the same wx import error. Michael, is this wx module
necessary? In the original Gnome-terminal dependent version of
relaxGUI (now attached to task #6847,
https://gna.org/task/index.php?6847), wx was not needed. Which
parts
of the GUI need this? Michael, I would suggest attacking the
problem
of integrating relaxGUI into the core of relax by looking at the
'relax' file in the base directory and adding the command line
options. Then once the Bieri GUI mode is selected, throw a
RelaxError
if wx is not installed (well, only if this is absolutely required)
so
that the error is clean and lets the user know that they need to
install the Python wx module. Then the GUI can be launched in a
similar manner as the prompt and script interfaces. Most of this
can
be achieved by code duplication and slight modification. To
generate
patches for me to apply to the branch, just type:
$ svn diff > patch
and then send the patch to the relax-devel mailing list.
Cheers,
Edward
2009/11/18 Sébastien Morin <sebastien.morin.1@xxxxxxxxx>:
Hi Michael,
First of all, thank you !
Second, I just tried the GUI and got an error related to
importing of
'wx'. On my Linux system (Gentoo Linux), 'wx' is provided through
the
program 'wxpython'. I think this should thus be added as a
dependency
with, possibly, some check to verify that it is present on the
system
used...
At first glance, the GUI seems very nice !
Regards,
Séb
Michael Bieri wrote:
Follow-up Comment #2, task #6847 (project relax):
Hi
I atatched the relaxGUI program files. To start the GUI type:
$ relax relaxGUI.py in the corresponding folder.
The Extra > Settings... dialog is not implemented yet, but I
can add
additional entries.
Let me know what you are thinking about and what could be
better.
Cheers
Michael
(file #7332)
_______________________________________________________
Additional Item Attachment:
File name: relaxGUI.tar.gz Size:381 KB
_______________________________________________________
Reply to this item at:
<http://gna.org/task/?6847>
_______________________________________________
Nachricht geschickt von/durch Gna!
http://gna.org/
_______________________________________________
relax (http://nmr-relax.com)
This is the relax-devel mailing list
relax-devel@xxxxxxx
To unsubscribe from this list, get a password
reminder, or change your subscription options,
visit the list information page at
https://mail.gna.org/listinfo/relax-devel
--
Sébastien Morin
PhD Student
S. Gagné NMR Laboratory
Université Laval & PROTEO
Québec, Canada
--
Sébastien Morin
PhD Student
S. Gagné NMR Laboratory
Université Laval & PROTEO
Québec, Canada
Re: [task #6847] The Bieri graphical user interface.