On Thursday 20 November 2003 11:00, jr wrote: > Hi Miguel, > if you simply reduce data you don't get compatible with RasMol script > language. And I am very unhappy with such a decision and I think all the > biomacromolecular structure people dealing with PDB files are.
I don't think Micheal said that Jmol is not going to support it at all... but only not on short terms... He's doing an amazing job, but cannot solve all the Rasmol/Chime problems in just a few months just by himself... (see below, when he mentions filing a bug...) > Parsing a > PDB header is a pain and not really necessary for a molecule display > applet, but parsing the structure data is prerequisite, so all features of > an ATOM or HETATM (CONECT, TER) should be represented in the Jmol atom > object, and they all should be addressable by script selection commands. How are the alternative conformations displayed in Rasmol? Not at the same time, or are they? If not, we could have Jmol read the alternatives into separate frames, so that one can select one conformation by selecting a specific frame... > > > restrict [Glu]71;A > > > would in RasMol select only one of them and not [Glu]71;B, but on the > > > command line, the ";" is not interpreted as alternative conformation > > > separator. > > > > Oh ... I don't feel very good ... I think I am going to be sick. > > > > The RasMol documentation I worked from ... 2.6 ... does not mention > > alternate locations. I just looked at the 2.7.2 documentation which was > > recently posted on openrasmol.org and they *do* specify ; as the > > character for selecting alternate locations. > > > > This, of course, conflicts with Chime's use of ; as the statement > > separator character. This is a problem. We have to make a choice here what we want to support... Rasmol 2.7.2 or Chime... I guess this cannot be done in a universal way... > > There is currently not a good mechanism in Jmol to support alternate > > locations. Therefore, I will modify the .pdb reader to take the first > > conformation it sees and discard the others. > > > > I have filed a bug report so that we don't lose the fact that we are not > > supporting alternate conformations. Right, and since it is *filed* as a bug, it is the intention that it *will* be solved. > > > picking the erroneously bound oxygen atoms, [Glu]71.OE1;A and > > > [Glu]71.OE1;B (in RasMol notation) > > > in the pdb1plc.ent file: > > > ATOM 1123 OE1AGLU 71 -3.634 19.815 16.036 0.50 28.03 > > > ATOM 1124 OE1BGLU 71 -3.850 18.852 17.241 0.50 27.59 > > > echoes (not in my TEXTAREA but) in the JavaConsole > > > notifyAtomPicked(1122,Atom: OE1 1123 GLU 71 Chain: ) > > > notifyAtomPicked(1123,Atom: OE1 1124 GLU 71 Chain: ) > > > there is no alternative location identifier > > > > There is not alternate location identifier because I didn't know such a > > thing existed :-) > > This is the moment to repeat my description of already existing atom > selecting primitives in RasMol 2.7.2 > > Primitive: [Gly]1:E.Ca;B/1 > Residue name: Gly, ATP, [SO4], As? > Residue number: 1,16,12,9-20 > Insertion code: not implemented! > Chain: :E > Atom name: *.Ca, *.C??? > Alternate location indicator: *;B (no wildcard) > Model: */1 (no wildcards but /0) > Wildcards: **:*.????/0 > > RasMol can deal with up to 5 separate molecules > molecule <number> this should be made a primitive, too How are molecules defined here? Are they separate overlaid molecules, or parts of a supramolecular system, e.g. in hemoglobin? > The Insertion code is really important and I'm very sad about the lack of a > primitive for it, but RasMol in contrary to Jmol is able to deal with it > (there must be an internal primitive). Try 1hag > http://www.imb-jena.de/cgi-bin/SCOPlnk3.pl?ASK=1hag > Jmol wouldn't load the molecule, and if, how do I select between > ATOM 20 CA GLY E 1F 20.593 14.133 18.600 1.00 50.00 1HAG 213 > ATOM 30 CA GLY E 1D 18.774 17.420 21.596 1.00 50.00 1HAG 223 How was it marked again to be two complementary, alternative locations? > I think we have to discuss about primitive notation. > > > > a rule may be, draw bonds in between each alternative conformation and > > > don't bind an atom from the other conformation if the same type is > > > already bound but at the borders > > > > Rather than do that, we are just going to take the first confirmation. > > I totally disagree. I'll have a look at the situation when I get Rasmol 2.7.2.1 compiled... and see how Jmol can deal with alternative conformations... Egon -- [EMAIL PROTECTED] PhD on Molecular Representation in Chemometrics Nijmegen University http://www.cac.sci.kun.nl/people/egonw/ GPG: 1024D/D6336BA6 ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ _______________________________________________ Jmol-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jmol-users
