Geoff said: >>I would like to restart the axis orientation discussion from the >>beginning.
> The restart is an *excellent* idea. Good >>We are going to talk about orientation first ... rotations will be a >> separate discussion. > > Yes, separating these *should* help to clarify matters. I had hoped so ... but I am beginning to wonder. >>Right Hand rule says that with the right hand: >> thumb -> +X >> index finger -> +Y >> middle finger -> +Z > >>*Someone please confirm this* > > I completely agree i.e. > > > +Y > | > | > | > - - - +X > / > / > / > +Z (pointing towards viewer) > > (this diagram uses a fixed width font and hopefully will not get > scrambled in email) Very good drawing! >>RasMol/Chime have the +X axis pointing to the right, the +Y axis > pointing >>down, and the +Z axis pointing away from the user. This is a valid >> right-hand system. > >>*Someone please confirm this* > > A qualified yes. > > The versions of RasMol and Chime that I have tested all display > molecules using the correct right-handed system. Presumably, the > 'biochemists' would have noticed if this was not the case and structures > did not display L-amino acids or right-handed alpha helices ;-) Good point! > However this is complicated by the fact that for Chime and RasMol 2.6 > beta 2a(Windows version) the 'set axes on' command draws the axes with > the Y label on the negative end giving the impression that the axes are > drawn with a left-handed system. I understand you to say that you believe that this is a *labelling* issue, not a change in orientation. That makes sense to me. It is hard to believe that a *simple* bug could lead to a complete change in orientation. > As I have mentioned previously, a (probably) related issue is that for > Chime and RasMol 2.6 beta2a (Windows version) 'set unitcell on' draws a > unit cell displaced in the -Y direction. i.e. the unit cell is not > 'wrong' just drawn in an unexpected location (at least for me). I don't really understand what you are saying here. But since it relates to unit cells, let not pursue it at this time. > Both the 'set axes on' and 'set unitcell on' options are corrected in > the Linux version of RasMol 2.7.2.1. i.e. molecules and axes are > displayed according to the right-handed system and unit cells are > displayed in the 'expected' place. This is in complete agreement with > the way the unitcell-enabled version of the Jmol applet works. Good. >>There is a switch on the popup menu called 'Axes RasMol/Chime'. This >> switch controls whether or not the +Y axis points down or up. By > default, >>this switch is 'on', for RasMol/Chime compatibility. That means that by >> default, the +Y axis points down. > >>If you uncheck this box, then the +Y axis points upwards. The molecule > is >>(of course) rotated 180 degrees to reflect this. > >>Whether this checkbox is checked or not, the Jmol axis orientation is >> always right-handed. > >>*Someone please confirm this* > > Yes I can fully confirm this. The switch does not affect the handedness > of the system but does change the orientation of the molecule (or > viewers' perspective) by 180 degrees. (This had confused me earlier as I > was expecting a change in handedness of the axes). Good. >>My confusion surrounding this issue is caused by the fact that >>RasMol/Chime has the +Y axis pointing downwards. I find this > orientation >>unnatural. > >>Q: In the abstract, independent of any existing applications/scripts > that >>may exist, what do people think about this? Do you believe that the +Y >> axis should point upwards or downwards? > > 'In the abstract', I am not too bothered about this. For me, a key part > of visualising 3D structures on 2D screens is manually rotating them > using the mouse. So as long as I can rotate the structure so that +Y > points up, +X points right and +Z points to the viewer I am happy. That > said, it seems reasonable to 'open' molecular displays with +Y pointing > up and +X pointing to the right as this is the familiar convention (for > 2D graphs etc). OK >>Q: Given that there *are* existing applications and tutorials, what do > you >>think we should do about this issue? > > I do not have a great deal of experience with these sorts of > RasMol/Chime scripts and tutorials. Given the inconsistencies and > variations in axis display noted above, I am not sure whether script > problems just relate to 'orientation' or to 'handedness'. Can anyone > clarify this? For all practical purposes, I have never written any scripts. However, here is an example that makes sense to me: I have a protein and I want to hilite a particular region. To do so I change the color of that region, then position the molecule so that the region of interes is most visible. If Jmol does not have the same initial orientation as Chime/RasMol, then my rotation commands are going to have to be different (although I am try to stay away from rotations ...) > I would prefer the default behaviour of Jmol to be as 'straightforward' > as possible. i.e. +X points to right. +Y points up, Z points out to the > user. I feel that script issues only involving 'orientation' should be > easy to deal with by changing the script (or options in Jmol). OK > Script > issues involving handedness, if any exist, should *not* be accommodated > with options in Jmol. I understand you to say that Jmol should not accomodate left-handed systems. Geoff, thanks for taking the time to respond to this. Miguel ------------------------------------------------------- SF.Net is sponsored by: Speed Start Your Linux Apps Now. Build and deploy apps & Web services for Linux with a free DVD software kit from IBM. Click Now! http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click _______________________________________________ Jmol-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jmol-users
