[phenixbb] TLS refinement

[Pavel Afonine]

Hi Reza,

this may mean several things. Perhaps TLS group selection isn't optimal 
or incorrect? phenix.find_tls_groups relies on B factors to annotate TLS 
groups. Were B factors meaningful (refined before running 
phenix.find_tls_groups)?

You can try weight optimization.

Values of TLS matrices cannot be interpreted directly, they can be 
positive or negative. This is discussed in great details here:

http://journals.iucr.org/d/issues/2015/08/00/rr5096/rr5096.pdf

You can check if TLS matrices are physically meaningful by running

phenix.tls_analysis model.pdb

If it runs all the way till completion - all is  good. Otherwise, in 
case of nonsensical TLS it will stop and tell what's wrong.

Perhaps you enabled TLS refinement too early (model isn't good yet)?

Pavel


On 4/30/17 05:30, Reza Khayat wrote:
>
> Hi,
>
>
> I'm having difficulty with performing TLS refinement on a highly 
> symmetric system. The difference between the R_work and R_free 
> increases with TLS refinement (16.5/21.4 changes to 14.4/22.2). The 
> TLS tensors for some of the group are not sensible (if I understand 
> the literature correctly) -there are some negative numbers for the T 
> and L tensors (see below). The TLS groups were identified with 
> phenix.find_tls_groups. Any suggestions would be appreciated.
>
>
> REMARK 3    SELECTION: chain 'A'and (resid 38  through 68)
>
> REMARK 3    ORIGIN FOR THE GROUP (A): 56.6459-23.5648-59.0695
>
> REMARK 3    T TENSOR
>
> REMARK 3      T11: 0.2105T22: 0.2895
>
> REMARK 3      T33: 0.0962T12: 0.0829
>
> REMARK 3      T13: 0.0473T23: 0.0007
>
> REMARK 3    L TENSOR
>
> REMARK 3      L11: 0.0446L22: 0.0120
>
> REMARK 3      L33: 0.0096L12: 0.0246
>
> REMARK 3      L13: 0.0209L23: 0.0106
>
> REMARK 3    S TENSOR
>
> REMARK 3      S11: 0.0228S12:  -0.0268S13:  -0.0023
>
> REMARK 3      S21:  -0.0507S22: 0.0526S23:  -0.0287
>
> REMARK 3      S31:  -0.0296S32:  -0.0137S33: 0.0534​
>
>
> The data set is complete at 2.3Å resolution and there are no strange 
> crystal phenotypes.
>
>
> Best wishes,
> Reza
>
>
> Reza Khayat, PhD
> Assistant Professor
> City College of New York
> Department of Chemistry
> New York, NY 10031
>
>
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe: phenixbb-leave at phenix-online.org

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20170502/d88d0165/attachment.htm>