New atom types and topology/parameter files

[Christoph]

Dear CP2K users,

I would simulate on the MM level an artificial test system with CP2K 
(artificial refering here to the atom types), using CHARMM 
topology/parameter files.

As a minimal example, I have prepared a geometry optimization for a 
3-atomic molecule, which I fail to get working for MM only simulations. The 
input files for two versions, 1 and 2, are attached.

When running version 1 I get:


 *******************************************************************************
 *   ___                                                                   
    *
 *  /   \                                                                   
   *
 * [ABORT]  Unknown element for KIND <A>. This problem can be fixed 
specifying *
 *  \___/  properly elements in PDB or specifying a KIND section or getting 
in *
 *    |                     touch with one of  the developers!             
    *
 *  O/|                                                                     
   *
 * /| |                                                                     
   *
 * / \                                                   
 topology_util.F:1271 *
 *******************************************************************************


 ===== Routine Calling Stack ===== 

            6 connectivity_control_check_element_name
            5 connectivity_control
            4 topology_control
            3 fist_init
            2 fist_create_force_env
            1 CP2K

Program received signal SIGABRT: Process abort signal.

Backtrace for this error:
#0  0x23458CD in _gfortrani_backtrace at backtrace.c:258
#1  0x2320A20 in _gfortrani_backtrace_handler at compile_options.c:129
#2  0x42DE5F in _L_unlock_13 at funlockfile.c:?
#3  0x42DD2B in raise at pt-raise.c:37
#4  0x23BF037 in abort at abort.c:78
#5  0x1E20288 in __message_passing_MOD_mp_abort at message_passing.F:1034
#6  0x1E74B87 in __base_hooks_MOD_cp_abort at base_hooks.F:68
#7  0xF33EE6 in __topology_util_MOD_check_subsys_element at 
topology_util.F:1275 (discriminator 15)
#8  0xC2AE9A in __topology_MOD_connectivity_control at topology.F:355
#9  0xC2BBEC in __topology_MOD_topology_control at topology.F:187
#10  0x7B561B in __cp_subsys_methods_MOD_cp_subsys_create at 
cp_subsys_methods.F:152
#11  0x141C867 in __fist_environment_MOD_fist_init at fist_environment.F:161
#12  0x137345F in __fist_main_MOD_fist_create_force_env at fist_main.F:97
#13  0xBA17C1 in __f77_interface_MOD_create_force_env at f77_interface.F:752
#14  0x437116 in __cp2k_runs_MOD_cp2k_run at cp2k_runs.F:276
#15  0x43AA9C in __cp2k_runs_MOD_run_input at cp2k_runs.F:1099
#16  0x4301DE in MAIN__ at cp2k.F:281



I guess one cannot specify new element types in CP2K via the input files 
only. In version 2 of my example files the atom type is changed to "C" 
(both in the PDB and the CP2K KIND input files), which should be known by 
CP2K. However, I get with when running version 2:

 *******************************************************************************
 *   ___                                                                   
    *
 *  /   \                                                                   
   *
 * [ABORT]                                                                 
    *
 *  \___/       A string type object was expected, found end of line, File: 
   *
 *    |                    '../system1.prm', Line: 13, Column: 6           
    *
 *  O/|                                                                     
   *
 * /| |                                                                     
   *
 * / \                                         
 input/cp_parser_methods.F:1253 *
 *******************************************************************************


 ===== Routine Calling Stack ===== 

            5 read_force_field_charmm
            4 force_field_control
            3 fist_init
            2 fist_create_force_env
            1 CP2K

Program received signal SIGABRT: Process abort signal.

Backtrace for this error:
#0  0x23458CD in _gfortrani_backtrace at backtrace.c:258
#1  0x2320A20 in _gfortrani_backtrace_handler at compile_options.c:129
#2  0x42DE5F in _L_unlock_13 at funlockfile.c:?
#3  0x42DD2B in raise at pt-raise.c:37
#4  0x23BF037 in abort at abort.c:78
#5  0x1E20288 in __message_passing_MOD_mp_abort at message_passing.F:1034
#6  0x1E74B87 in __base_hooks_MOD_cp_abort at base_hooks.F:68
#7  0x19C149C in __cp_parser_methods_MOD_parser_get_string at 
cp_parser_methods.F:1255 (discriminator 2)
#8  0x15789FF in __force_fields_ext_MOD_read_force_field_charmm at 
force_fields_ext.F:434
#9  0x141D28D in __force_fields_MOD_force_field_control at 
force_fields.F:158
#10  0x141C90C in __fist_environment_MOD_fist_init at fist_environment.F:169
#11  0x137345F in __fist_main_MOD_fist_create_force_env at fist_main.F:97
#12  0xBA17C1 in __f77_interface_MOD_create_force_env at f77_interface.F:752
#13  0x437116 in __cp2k_runs_MOD_cp2k_run at cp2k_runs.F:276
#14  0x43AA9C in __cp2k_runs_MOD_run_input at cp2k_runs.F:1099
#15  0x4301DE in MAIN__ at cp2k.F:281
/home/jadzia/Dropbox/Software/installed/bin/cp2k: line 5:  2732 Aborted     
            ${script_dir}/cp2k.ssmp $@



If possible I would like to get this working with them CHARMM files rather 
than with CP2K keywords. I know that there are other programs which are 
specialized on MM simulations, but I want to use CP2K due to some special 
features it has. 

Looking forward to your comments.

Best wishes,
Christoph
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20170626/1199a909/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: version-1.tar.gz
Type: application/octet-stream
Size: 1794 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20170626/1199a909/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: version-2.tar.gz
Type: application/octet-stream
Size: 1791 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20170626/1199a909/attachment-0001.obj>