Re: [ccp4bb]: Another refmac/tlsanl TLS question

[Ethan Merritt <merritt@u.washington.edu>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

On Friday 07 June 2002 15:21, Ethan Merritt wrote:
> ***  For details on how to be removed from this list visit the  ***
> ***          CCP4 home page http://www.ccp4.ac.uk         ***
>
> Another thing I don't quite understand...

The light dawns.  I think I understand after all.
Please correct me if I'm on the wrong track.

The Biso values in the refmac5 are the raw values refined 
subsequent to the TLS model. They are independent of the
displacements described by the TLS, which have not yet
been applied (this was what I misunderstood originally).

Tlsanl converts the Biso and the TLS parameters jointly 
into a set of ADPs.  Yes that's what the documentation
says, but somehow I had the idea originally that refmac
had already combined them isotropically, and tlsanl was
just correcting that to an anisotropic description.

My other question (about overlapping TLS groups) 
still stands, however.

	thanks for any discussion,

		Ethan

>
> The Biso values in the pdb file output by refmac5 after TLS
> refinement are quite different from the Biso values output by tlsanl
> after converting the TLS description into individual ADP records
> starting from that very same refmac5 pdb file.
>
> I realize that the Uij parameterization is only an approximation to
> the actual atomic scattering described by Biso+TLS, but I would
> expect the Uij parameterization to be chosen so that _something_
> came out the same, and Biso (a.k.a. Beffective) is the logical
> quantity to hold constant.
>
> - What exactly is the quantity stored in the Biso field of ATOM
> records output by refmac5 after TLS refinement?
>
> - What is being fit (or minimized?) by tlsanl when it converts the
>   refmac5 output into a description of individual atomic ADPs?
>
> - If I shouldn't expect Biso to be the same before and after running
>   tlsanl, then what should I expect?  Should it get larger? smaller?
>   Does the direction or magnitude of change tell me anything about
>   the nature of the TLS model?

-- 
Ethan A Merritt       merritt@u.washington.edu
Biomolecular Structure Center Box 357742
University of Washington, Seattle, WA 98195
phone: (206)543-1421
FAX:   (206)685-7002