[phenixbb] refine : a priori TLS tensor eigenvalues

Pavel Afonine pafonine at lbl.gov
Thu Oct 21 08:45:40 PDT 2010

  Hi Bryan,

> thanks, phenix.tls looks helpful.

oh, good, I'm glad you find it helpful!

> .... and if the input pdb has derived anisou/B's, could this
> potentially interfere in the TLS protocol? i.e. would it be wiser to
> simply reset them every time?

It shouldn't really matter for phenix.refine. Probably it's better if 
you feed in a PDB file with total B-factors in rather than residual, but 
honestly I wouldn't expect any significant difference. Of course, if you 
want to have a rock solid answer then just try using both approached and 
compare the difference.

> i know ADP's/Bisos are derived from TLS matrices,

Only TLS component is derived from TLS matrices. This is explained here:


see article "Atomic Displacement Parameters (ADPs), their 
parametrization and refinement in PHENIX".


More information about the phenixbb mailing list