I presume you tried P32 21 also? Pointless cannot distinguish the
difference..
Refinement at 3A is tricky - you can overfit quite easily and maybe
Rfree is correctly flagging that there are still errors.
Can you see features which need rebuilding?
Eleanor
Emmanuel Prata wrote:
Dear all,
I have a structure (with 6 dissulphide bridges) at fairly low
resolution (3A) that I am trying to refine with Refmac. I've used
phaser with a template with 72% sequence identity. RFZ was 3.4 and TFZ
6.6, without clashes, with 80.7% of completeness. Space group was
p3121 (pointless) and Rmerge 0.12. When i tried refmac, Rfac drop to
24%, but Rfree only drop to 41%.
Can anybody explain this fact?
Thanks in advance,
Emmanuel Prata de Souza