Jason (and PyMOLers),

To clarify, B.dcd matches the number of atoms in B.pdb and A.dcd also matches the number of atoms in A.pdb.  In the previous example, if I perform steps 3 and 4 first then it works in loading B.dcd into object "B".  Then, if I try to open up A.pdb into a separate PyMOL window, loading of A.dcd into object "A" fails with the same error (obviously, with the atom numbers reversed).  Thus, somehow, the second PyMOL window is retaining/obtaining information from the first PyMOL window.

Okay, I've managed to make a little bit more progress on identifying this problem.  After navigating to the directory for A.pdb (/path/to/A/), I open A.pdb and load in A.dcd (the window is kept open).  No problem.  Then, on the command line (not in PyMOL), I navigate to the directory for B.pdb (/path/to/B/) and open B.pdb in a new PyMOL window (i.e. "pymol B.pdb").  In this (B) window, if I do a "pwd" in the PyMOL GUI, instead of saying "/path/to/B" it returns "/path/to/A"!  For some reason, it doesn't recognize that I've changed directories.  This explains why it was throwing the "atoms in object did not equal number of atoms" error because it was indeed trying to load A.dcd into object "B" (which have different atom numbers).  I assumed that navigating to "/path/to/B" on the command line (outside of PyMOL) and then calling PyMOL from there was enough for PyMOL to know where it was.  However, it appears that having one instance of PyMOL already open will somehow affect the path of the second window.  Perhaps there's a workaround for this?  I've never encountered this behaviour on Linux before.  

For now, to get around this, in the second window, I have to re-navigate to the right directory within MacPyMOL as the command line is not enough.

Sean

> From: jason.vertrees@schrodinger.com
> Date: Wed, 1 Feb 2012 09:40:20 -0500
> Subject: Re: [PyMOL] MacPyMOL Knowledge of Opened Structures
> To: magicmen@hotmail.com
> CC: pymol-users@lists.sourceforge.net
>
> Hi Sean,
>
> Sorry to hear you're having problems.
>
> > 1) Load A.pdb into MacPyMOL via "pymol A.pdb"
> > 2) Load a CHARMM simulation trajectory into object "A" via "load A.dcd, A"
> >
> > 3) Load a completely different structure (must have different number of
> > atoms than A.pdb), B.pdb, into a separate/new instance/window of MacPyMOL
> > via "pymol B.pdb"
> > 4) Load a CHARMM simulation trajectory into object "B" via "load B.dcd, B"
> >
> > Loading of this second trajectory complains with the following error:
> >
> > ObjectMolecule: plugin 'dcd' cannot open file because the number of atoms in
> > the object (4551) did not equal the number of atoms in the 'dcd' (5889)
> > file.
>
> I wrote this error message. PyMOL counts the atoms in the object and
> makes sure each frame of the trajectory has an equivalent number. This
> stops PyMOLers from accidentally working with incorrectly paired data
> -- which happened quite a bit.
>
> > I should mention that in the real example, "A.pdb" and "B.pdb" actually have
> > the same name but are located in different directories (in case it matters).
>
> Seeing this makes me think of two possibilities. First, you possibly
> made a simple mistake on disk and B.pdb and B.dcd don't actually
> match. Do these _exact_ files using the exact same commands work the
> same in Linux? (Can you md5sum the files to check?) Second, if PyMOL
> is sharing data then, because they have the same file name, you'll
> have to rename one to something else, like "A2".
>
> Let us know.
>
> Cheers,
>
> -- Jason
>
>
>
> >
> > In this error, the number "5889" corresponds to the number of atoms in A.pdb
> > while "4551" corresponds to the number of atoms in B.pdb.  Again, these are
> > two separate windows of MacPyMOL (i.e. I did NOT load A.pdb and B.pdb into
> > the same window).  Somehow, the first instance of PyMOL sees what is
> > happening in the other.
> >
> > I should mention that in the real example, "A.pdb" and "B.pdb" actually have
> > the same name but are located in different directories (in case it matters).
> >
> > On a non-Mac machine, these commands are fine.  As an "experienced" PyMOL
> > user, it was quite embarrassing when a set of commands/steps that worked on
> > a Linux workstation does not work on a Mac (especially when I was trying to
> > show the simulations to a VMD user).  I'm fairly new to Macs so maybe
> > there's something that I need to do?
> >
> > Also, when I load up B.pdb (the second structure), the MacPyMOL doesn't seem
> > to run "hide all" and "show cartoon" from my .pymolrc.
> >
> > Any help/suggestions would be greatly appreciated!
> >
> > Sean
> >
> > ------------------------------------------------------------------------------
> > Keep Your Developer Skills Current with LearnDevNow!
> > The most comprehensive online learning library for Microsoft developers
> > is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
> > Metro Style Apps, more. Free future releases when you subscribe now!
> > http://p.sf.net/sfu/learndevnow-d2d
> > _______________________________________________
> > PyMOL-users mailing list (PyMOL-users@lists.sourceforge.net)
> > Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users
> > Archives: http://www.mail-archive.com/pymol-users@lists.sourceforge.net
>
>
>
> --
> Jason Vertrees, PhD
> PyMOL Product Manager
> Schrodinger, LLC
>
> (e) Jason.Vertrees@schrodinger.com
> (o) +1 (603) 374-7120