search

ReliaSolve / cctbx_project

Computational Crystallography Toolbox
https://cctbx.github.io
Other
0 stars 0 forks source link

Run the four PDB files Jane sent through cctbx and Reduce H placement #9

Closed russell-taylor closed 4 years ago

russell-taylor commented 4 years ago

Run them with no optimization and compare. Run them with optimization done on both and compare. Send diffs to Jane so that she can see what happened with each of them.

russell-taylor commented 4 years ago

Ran into problems using the low-level read routines from Vincent's test script. Dorothee recommended using the higher-level model manager interface, which was able to load at least one of the files.

Got the same error ("Cannot find CCP4 monomer library. Please define one of these environment variables: MMTBX_CCP4_MONOMER_LIB, CLIBD_MON") when calling mmtbx.hydrogens.add(), so I guess I just chased it deeper into the library.

russell-taylor commented 4 years ago

Was able to get the required files from the MolProbity installation so can now produce models within cctbx.

Will look at running Dorothee's new add() function and think about how to generate output from it and Reduce that can be compared one with another. The easiest approach is probably to generate two PDB files (or a CIF and PDB file) and then run a 3D viewer to look at them one at a time/overlapped.

russell-taylor commented 4 years ago

The cctbx hygrogenate.py script changes the atom numbers on the original file and also strips off a lot of header data, making text diff comparisons on its output vs. Reduce output problematic. I'll take a look at the models in coot, which is how Dorothee was comparing them when she demonstrated hydrogenate, to see if that gives us what we need.

russell-taylor commented 4 years ago

Hydrogenate did not complete on three of the four models. Packed up the results (including the one that they both worked on) and uploaded to the shared drive that Dorothee made, then informed the team via email.