[phenixbb] Can not decrease R-free value

Pavel Afonine pafonine at lbl.gov
Sun May 26 06:52:50 PDT 2019


from your description you seem to be doing the right things but the 
Rfree-Rwork gap is indeed rather large than expected at this resolution. 
If you send me model and data (off list) then there is a good chance I 
figure out what's happening and offer a  solution (if any ligand cif 
files involved -- send them too).


On 5/24/19 08:51, Pham, Hang wrote:
> Dear all,
> I am refining my structures at 2.5 A. The model is well fitting in the 
> electron density. I already add water and ligands.
> I’m trying these options (XYZ coordinates, real-space, TLS parameters, 
> Individual B-factors, Optimize X-ray/stereochemistry weight, Optimize 
> X-ray/ ADP weight, Reference model restraints, Use secondary structure 
> restraints, Simulated annealing (Cartesian), Automatically correct 
> N/Q/H errors) to refine the structure but the R-free still be high.
> The reference model is 98% identity with the structure that I’m 
> building. After refining, the R-work always decrease while R-free 
> increase. My last values for R-work is 0.2501 and for  R-free is 0.3246.
> I’m lost and don’t know what to try more.
> Can anyone could explain me what happens here ? What others options 
> that I could use to optimize R-free value ?
> Thank you in advance !
> Hang Pham

More information about the phenixbb mailing list