Hello CCPN-team!
I have used Analysis for quite a while now doing "real stuff". I
really like the program and appriciate the speed at which you develop
new features and fix bugs!
Using Analysis I have found a few oddities, which may or may not be
bugs. I am running Analysis (ver 1.0.4) under Mac OS X 10.4.1. I use
tk and tcl ver. 8.4 and Python 2.3.
General:
Some windows seem to have default size zero, especially this seems to
be true for windows handling files (like the Open project window).
They just appear as a very small rectangle showing only the Mac
"Close/Minimize/Expand" buttons.
Strip related stuff:
Adding a new strip makes the new strip active, but using "navigate
to" from another window sets the postition in both the new strip and
the last active strip. This is also true for other commands that
should go to just the active strip, like changing z-position using
the slider.
Using the "Find peak" button in the "Peak selection" Panel always
finds the peak in the first strip and not in the currently active.
The "Remove" strip button in the spectrum window does not remove the
active strip, but the last strip.
If the an active strip is removed, the new active strip gets the z-
position of the removed strip. (But the shown position is not altered
until the "slider" is moved.)
I also have a "wish list":
It would be great if the assignment panel had an option to show only
possible assignments for which the sequence number is the same in two
or more dimensions. This would for example help if a 3D 15N-NOESY
spectrum is to be assigned, given a known HN and N assignment. Since
these resonances always come in pairs it would be great to be able to
remove all assignemt suggestions in which the HN and N belongs to
different residues in one of the H dimensions and in the N-dimension.
I would like a feature in which a known assignment and structure
could be used to assign a NOESY-spectrum. Ansig for Windows has a
feature in which all distances up to a given distance can be plotted
in the NOESY spectrum as circles (larger circle = shorter distance)
with the assignments. This makes it easy to make and verify
assignemts in the spectrum.
Something which probably is alread on your "to do" list is to read
structures and compare them to constraint lists, make statistics and
then produce a peak list with all violations. It would be very useful
in stucture determinatinos!
Thanks for great software!
Magnus Helgstrand
|