Hi,

I was trying to troubleshoot this and it seems to be a problem related to the FEM map only. When I open my FEM.mtz and do RSR in coot it works well if I choose the 2mFoDfc map but not the FEM map. I'm still not sure why it stopped working since coot was behaving normally just yesterday. Maybe it had something to do with the cif file I tried to import earlier (see previous message).

Yazan

PS I'm replying to my previously sent message, I hope it ends up in the same thread...

On Thu, Oct 16, 2014 at 11:41 AM, Yazan Abbas <yazan.abbas@mail.mcgill.ca> wrote:

Hi,

I am encountering a problem with coot real space refinement (RSR). Basically when I try to fit any residues using RSR the atoms come apart and fit into the center of the density, possibly because it is not reading in any geometry restraints. This happens for examples when I am opening my FEM .mtz map (either manually or 'open in coot' via phenix).

The only times coot RSR behaves normally is when I load it via phenix after running phenix refine, or when I open coot via the command line and auto-open an mtz (the same one generated by phenix refine). Also, fetching a pdb and map via EDA/PDBredo still allows me to use RSR.

When I encountered the problem I tried to reinstall coot 0.8 stable and also tried to install the latest pre (0.8.1-5404), and upgraded to phenix 1.9-1692. My coot installation comes from the stand alone packages of the Scott lab website.

The only thing I can think of as having caused this is when I was trying to fit an unusual ligand into my density, I used phenix elbow to generate a cif file and then loaded that into coot (--> import cif dictionary). Shortly afterwards coot gave up on real space refinement under most circumstances.

Have I somehow replaced a default dictionary file?

Thanks,

Yazan

--
PhD Candidate
Dr. Bhushan Nagar Lab
McGill University
Tel: (514)-398-7271
http://bhushan-nagar.lab.mcgill.ca/

PhD Candidate
Dr. Bhushan Nagar Lab
McGill University
Tel: (514)-398-7271
http://bhushan-nagar.lab.mcgill.ca/