On 28 April 2010 00:50, Michael Bieri <michael.bieri@xxxxxxxxxxxxx> wrote:
I like the tree view idea. Do you think of a a collection of amino acids
(dna, rna...) that is initially linked to the sequence or should the
entries be read from the pdb file.
I was thinking of something more like:
+ Chain A
+ Chain B
+ Drug candidate 7385
This is the molecule level. Then after clicking on the plus you have
something like:
- Chain A
+ 1 Ala
+ 2 Trp
+ 3 Gly
+ Chain B
- Drug candidate 7385
+ UNK
This is the residue level. Then on the next level, the spins:
- Chain A
+ 1 Ala
- 2 Trp
N
NE1
+ 3 Gly
+ Chain B
- Drug candidate 7385
- UNK
C1
C2
C3
H1
H7
N5
N10
This is the structure of the molecule-residue-spin data containers in
relax, ccpn, bmrb, pdb, etc., just put into a tree view. My idea is
that if you click on NE1, then to the right appears the total
information content for that spin in another GUI element.
I also think reading sequence from pdb file is by far the best and most
appropriate method! This will never get disabled. Maybe, we should
introduce a correction function that converts (or adjusts) assignments
to the pdb file, if the assignment was not done with the same numbering?
Ok, I wasn't sure if you wanted to replace this. As for adjusting
sequence information, this is very difficult due to insertions and
deletions. Point mutations are ok, you just don't load in the residue
names. Also relax comes with many user functions for renaming and
renumbering spins/residues/molecules, so maybe the future GUI can use
these. Maybe a right-click in the tree view could bring up these
options.
Cheers
Edward d'Auvergne wrote:
On 28 April 2010 00:13, Michael Bieri <michael.bieri@xxxxxxxxxxxxx> wrote:
Yes, I think we should make a dialog that already displays the sequence,
so the users are able to see if the sequence is correctly read in. I
will add this element at a later step.
This would be good for a future version. What do you think of that
tree view GUI element to display the hierarchical structure? This
could be placed in its own class so that it can be reused inside a
frame/window in the future, so that you click on the spin and all the
info is displayed in another GUI element next to it.
What do you think? I even would prefer to have a text box, where users
can paste a sequence in, such as in CcpNmr.
Users are fussy and have their own way of doing things. Why have the
sequence typed out if you can read it directly from a PDB file? We
shouldn't artificially restrict user's abilities as the philosophy
behind the design of relax is flexibility, which hopefully can be
transported into the GUI. Pasting a sequence in would be quite useful
for some users. But if you would like this to be the only choice,
you'll first have to write the sequence data for this thing
http://en.wikipedia.org/wiki/Paclitaxel ;) Anyway, any way of loading
the sequence data will be accepted into the core of relax and can
accessed via any UI element - that would just add more flexibility
which is a good thing. I do think that reading the sequence from PDB,
a specific columnar formatted sequence file (can be any data file
which contains the sequence info in certain columns), or pasted
sequence data should all be accessible as ways of filling the
molecule-residue-spin data container structure. Just don't forget the
RNA/DNA/sugar/small molecule people - you never know, you might be
working on one of these (maybe complexed with a protein) in the
future!
Regards,
Edward
_______________________________________________
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
Re: r11106 - /branches/bieri_gui/gui_bieri/analyses/auto_rx_base.py