Hi Pavel,

I made a mistake with the output from phenix.reduce, now it works well. In my first attempt, the output from phenix.reduce was a file whose only content was: "USER  MOD reduce.3.24.130724 H: found=0, std=0, add=0, rem=0, adj=0". It was solved when I renamed the input pdb. I took erroneously the first file as an input for phenix.refine and it produced an error (obviously). Now I repeated the refinement following exactly your instructions. I got the same results (as expected).

version: dev-2247

Final: r_work = 0.1755 r_free = 0.2044 bonds = 0.030 angles = 0.598

version: 1.9-1692

Final: r_work = 0.1663 r_free = 0.2017 bonds = 0.011 angles = 1.321

El Martes, 15 de diciembre, 2015 22:07:15, Pavel Afonine <pafonine@lbl.gov> escribió:

Hi Ezequiel

phenix.reduce doesn't work well in my hands. I used "automatically add H's" in the phenix GUI instead.

"automatically add H's" in the phenix GUI uses phenix.reduce to add H. The machinery that actually add H is exactly the same.

PS: phenix.reduce reads my file, but it output a pdb file with the added H's with "0" in the second column (and "new" in the last column)

Does this create any problem?

Pavel