[phenixbb] Cryo_fit execution data test GTPases fails

[CUNIASSE Philippe]

Dear all,
After installing cryo_fit in phenix and gromacs_cryofit on my Macbook Pro M1, I now try to test the code using the example data GTPase.

The job encounters immediately a problem but the GUI suggests that the job is still running.

The log suggest that the first step did not run successfully (below). I presume setting the topology of the protein but I don’t know.
Another question is how is it possible to set other parameters for cryo_fit (like ff…) in the GUI or should I run it in command line ?

Thank you for your help and suggestions.
Philippe.



       User's computer's operating system: Darwin







###############################################################################################

#################### Step 1: Make gro and topology file by regular gromacs ####################

###############################################################################################

       Current working directory: /Applications/phenix-1.20.1-4487/modules/cryo_fit/tutorial_input_files/CryoFit_1/steps/1_make_gro

       command:  python 1_before_pdb2gmx_prepare_pdb.py GTPase_activation_center_tutorial.pdb 0 0 0 True

       Current working directory: /Applications/phenix-1.20.1-4487/modules/cryo_fit/tutorial_input_files/CryoFit_1

             This pdb file,  /Applications/phenix-1.20.1-4487/modules/cryo_fit/tutorial_input_files/GTPase_activation_center_tutorial.pdb , has  6489  atoms

       Approximately, for this number of atoms, one 3.1 GHz Intel Core i7 took 7 seconds to make a gro file.



       command:  python 2_runme_make_gro.py /Applications/phenix-1.20.1-4487/modules/cryo_fit/ amber03 True True /usr/local/cryo_fit/cryo_fit/bin//

Step 1 didn't run successfully



User's command was

phenix.cryo_fit /Applications/phenix-1.20.1-4487/modules/cryo_fit/tutorial_input_files/.phenix/project_data/run_1.eff





phenix.cryo_fit alone without any arguments introduces full options.



If a user sees a message like

    "Fatal error:

    Atom xx in residue xx xxx was not found in rtp entry xx with xx atoms

    while sorting atoms."



    or

    "Fatal error:

    Residue xxx not found in residue topology database"



    Solution if these residue/atoms are important:



       Fix wrong names of atoms/residues. Running real_space_refine via phenix GUI will show which atoms need to be fixed.

       If gromacs amber03 force field doesn't have parameters for these residue/atoms, you may need to add appropriate parameters.

       If you added parameters, please email me (doonam.kim at pnnl.gov), I want to recognize your contribution publicly.

       Most MD simulation force fields do not support all kinds of rare residue/atoms.

       cryo_fit2 is under development to address this issue using phenix.eLBOW



    Solution if these residue/atoms are not important:



       Remove these "wrong" atoms/residues from user's input pdb file. Run cryo_fit again.



    Solution if user's input pdb file is big:



       Probably cryo_fit will change conformation just minimally, I would extract out these "wrong" atoms/residues from user's input pdb file, then add these extracted lines to cryo_fitted file later.





Email phenixbb at phenix-online.org or doonam.kim at pnnl.gov for any feature request/help.




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20230504/52b4bc9e/attachment.htm>