Hi, - what if you do the exact same refinement but without SA? Do you still see the same behavior of R-factors? - does your input PDB file contain any ANISOU records? - if you send me the data and model (and the exact command you used), then I will be able to tell what exactly happens. Pavel.

I am doing refinement using phenix. I did simulated annealing, Rfree goes up to 10% more than before. (40%) and R stays at 35%. If use previous phenix refinment file for same pdb and data , R and Rfree is 25/27%. I thought even different r-free set after simulated annealing and 7 cycles of refinement in phenix should bring Rfree down, but not ?

But same data set and model , in CNS, R/Rfree 26/28%.

Any suggestions will be appreciated.