Hi Ralf, in fact, phenix.refine outputs information about refined f' and f'' in ready-to-PDB-deposit format under RAMARK 3 prefix. For eample: REMARK 3 REMARK 3 ANOMALOUS SCATTERER GROUPS DETAILS. REMARK 3 NUMBER OF ANOMALOUS SCATTERER GROUPS : 2 REMARK 3 ANOMALOUS SCATTERER GROUP : 1 REMARK 3 SELECTION: name BR REMARK 3 fp : -0.8792 REMARK 3 fdp : 5.8029 REMARK 3 ANOMALOUS SCATTERER GROUP : 2 REMARK 3 SELECTION: name CO REMARK 3 fp : 0.0000 REMARK 3 fdp : 0.0000 REMARK 3 I hope this information is preserved when such structures get deposited into PDB -:) Pavel. On 1/20/10 9:44 PM, Ralf W. Grosse-Kunstleve wrote:

...

Do I understand correctly that there is no way to force refine to skip the serial in the output pdb file name, i.e. foo.pdb, not foo_001.pdb?

Yes, that's true. The main reason is that the same run writes foo_001.eff foo_002.def When I implemented this I thought the serial numbers are the only practical way to keep things straight. Later we added storage of refined f' and f" to the .def file (since there is no place in the .pdb file that would survive passes through external programs). Because of this the serial numbers seem even more important to me. Ralf _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

Dear phenix experts: I have a data in R32 at 2.3A. 4 molecules/A.U. I got the following error message when I run refinement wiht phenix v1.5-2: =============================== refinement start ============================== Traceback (most recent call last): File "/usr/local/phenix-1.5-2/phenix/phenix/command_line/refine.py", line 11, in <module> command_line.run(command_name="phenix.refine", args=sys.argv[1:]) File "/usr/local/phenix-1.5-2/phenix/phenix/refinement/command_line.py", line 89, in run call_back_handler=call_back_handler) File "/usr/local/phenix-1.5-2/phenix/phenix/refinement/driver.py", line 1121, in run call_back_handler = call_back_handler) File "/usr/local/phenix-1.5-2/phenix/phenix/refinement/strategies.py", line 185, in refinement_machine log = log) File "/usr/local/phenix-1.5-2/cctbx_project/mmtbx/utils.py", line 1371, in fmodel_manager max_number_of_bins = max_number_of_resolution_bins) File "/usr/local/phenix-1.5-2/cctbx_project/mmtbx/f_model.py", line 358, in __init__ mask_params = self.mask_params) File "/usr/local/phenix-1.5-2/cctbx_project/mmtbx/masks.py", line 175, in __init__ self.compute_f_mask(twin = twin) File "/usr/local/phenix-1.5-2/cctbx_project/mmtbx/masks.py", line 264, in compute_f_mask asu_mask.compute(sites_frac, atom_radii) RuntimeError: mmtbx Error: the mask grid size: 576 * 576 * 270 = 89579520 is too big maximum allowed size is 67108863 64 bit OS and software might be required for such a big mask. How can I tackle this problem? Any suggestion is highly appreciated. Xue

but not reflection_file_tools.htm

I'll fix it when I get a chance. Thanks for pointing out the oversight. Ralf