[phenixbb] Multiple TLS groups per molecule

[Nathaniel Echols]

On Tue, Nov 15, 2011 at 7:27 AM, Yarrow Madrona <amadrona at uci.edu> wrote:
> One more question.  When you refine in subsequent rounds I read that
> phenix.refine will refine as anisotropic because of the ANISOU labels
> attached to the PDB. Is it necessary to enter something like
> "stragegy=tls+isotropic"? I am just worried that if I explicitly ask for
> individual isotropic refinement this might occur without tls refinement.

This is only a problem at relatively high resolution - the key parameter is

    switch_to_isotropic_high_res_limit = 1.5
      .type = float
      .help = If the resolution is lower than this limit, all atoms selected \
              for individual ADP refinement and not participating in TLS \
              groups will be automatically converted to isotropic, whether or \
              not ANISOU records are present in the input PDB file.
      .short_caption = Resolution limit for automatic anisotropic ADPs

(".short_caption" is what appears in the GUI)

So if your resolution is worse than 1.5, I think it will do what you
want automatically.  Otherwise, you can avoid the problem by
converting your PDB file to isotropic atoms
(convert_to_isotropic=True, I think - phenix.pdbtools can also do
this), and/or specify refine.adp.isotropic="not (element H or element
D)".

-Nat