Hi David,
this is bug, indeed. I also hit this some time ago but didn't report it
so far. The actual problem is that the cmd.super('1acb', '426d') fails
completely and only aligns a single atom, resulting in a Null-matrix
that collapses all of 1acb into a single point. The RMSD of 2.8 that you
get in your last call is just an artefact that arises from aligning the
correct protein to the collapsed one. Just run your example with GUI and
you see what happens.
fetch 1acb
fetch 426d
copy 1acb_2, 1acb
super 1acb, 426d
super 1acb, 1acb_2, object=aln
zoom aln
I think the super command should check if the resulting matrix is
orthonormal or at least not a Null-matrix.
Cheers,
Thomas
On Wed, 2010-09-22 at 18:28 -0400, David Hall wrote:
> In preparing my test for the odd request that I just sent, I came
> across what seems to be a bug in super. It has to do with this weird
> nucleic acid thing I was talking about.
>
> Here's the script this time:
> from pymol import cmd
>
> cmd.fetch('1acb', async=0)
> cmd.fetch('1acb', '1acb_2', async=0)
> cmd.fetch('426d', async=0)
>
>
> print "1acb_2"; print cmd.super('1acb', '1acb_2')[0]
> print "426d"; print cmd.super('1acb', '426d')[0]
> print "1acb_2_again"; print cmd.super('1acb', '1acb_2')[0]
>
> ~> ~/src/pymol_trunk_20100922/pymol -qrkc script2.py
> PyMOL>run script2.py,main
> 1acb_2
> 0.0
> 426d
> 0.0
> 1acb_2_again
> 2.80655193329
>
> super should give an rms between 1acb and 1acb_2 of 0.0, which it does
> at first. But once you super your protein to a nucleic acid that has
> a calcium ion, when you run exactly the same command, you all of a
> sudden get 2.8. Very weird...
>
> Note, this does not happen if you use 100d, which is a nucleic acid
> without a calcium ion, does not happen if you use a pdb file with a
> calcium ion, and does happen with 1d56, which is another nucleic acid
> with a calcium ion I just pulled out.
>
> In pymol 1.2, it also behaved incorrectly:
> ~> pymol -qrkc script2.py
> PyMOL>run script2.py,main
> 1acb_2
> 0.0
> 426d
> nan
> 1acb_2_again
> nan
>
>
> -David
--
Thomas Holder
Group of Steffen Schmidt
Department of Biochemistry
MPI for Developmental Biology
Spemannstr. 35
D-72076 Tübingen
------------------------------------------------------------------------------
Start uncovering the many advantages of virtual appliances
and start using them to simplify application deployment and
accelerate your shift to cloud computing.
http://p.sf.net/sfu/novell-sfdev2dev
_______________________________________________
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
