Step 2. Convert PDB file to LAMMPS input file
In the previous step, we prepared the input PDB files either in CG or AA (step5_assembly.pdb) format.
In the "AA PDB file -> CG PDB file" section, we will introduce a command of the spica-tools,
map2cg,
to convert AA PDB file to CG PDB file for users building AA PDB files.
If you already have CG PDB files, you may skip this section.
Since LAMMPS reads its own formatted data file, it is required to convert the PDB file to LAMMPS data file.
To convert CG PDB file, we provide another command of the spica-tools, called
setup_lmp.
In the "CG PDB file -> LAMMPS input file" section, we will introduce how to use setup_lmp.
* The commands start with "$" and "%" in the gray source code box are for bash shell
(or any linux shell) and TCL shell on VMD, respectively.
AA PDB file -> CG PDB file
We use the command-line tool
cg_spica to convert the AA PDB file to CG PDB file,
as mentioned in
the first page of the tutorial.
This tool includes a CG mapping command
map2cg,
which is to convert AA PDB to CG PDB based on mapping information described in a JSON file
(for details to its document). For typical lipids, thier mapping information is already written
in the JSON file. So, for the target system containing DOPC lipids and water, we just excute
the following command to map the AA PDB to CG PDB:
$ cg_spica map2cg step5_assembly.pdb final.pdb
The first and the second arguments following "map2cg" are input AA PDB and output CG PDB, respectively.
Please make sure that the console finally outputs "Normal termination" without any errors
after invoking the command. The obtained CG PDB
final.pdb will be used to generate LAMMPS
input files in the next section.
CG PDB file -> LAMMPS input file
Before using setup_lmp,
we need to add the information manually with an unix command because the PDB file generated
with CHARMM-GUI does not have the simulation box information. The command should
be like:
$ sed -i 1i"CRYST1 66.790 66.790 79.000 90.00 90.00 90.00" final.pdb
The simulation box size is written in the
step5_assembly.str file included
in the generated CHARMM-GUI top directory.
To execute
setup_lmp,
one should provide the topology and number of molecules, database file, and input PDB file
as arguments. For example, if the PDB file, named
final.pdb,
contains 128 DOPC lipids and 3000 water molecules, the command will be:
$ cg_spica setup_lmp DOPC.top 128 WAT.top 3000 spica_db.prm final.pdb
where the
DOPC.top and
WAT.top contain the topology of DOPC and water molecule,
respectively.
spica_db.prm is the force field file we provide.
final.pdb is the PDB file contains
the intial CG coordinates.
One should be careful with the order of arguments.
Especially, molecule topology and number should be the same as in the PDB file.
For the topology files, they are included in the force field file (Check
topology/*.top
included in
spica_v1.1.tar.gz
).
Other topology files can be generated with another spica-tool command
json2top,
based on mapping information in the JSON file (for details to its document).
If the conversion is successful, setup_lmp will generate three output files,
- DATA.FILE
- PARM.FILE
- out.psf
The input script will read these files.
The contents of DATA.FILE can be visualized on VMD with
Topotools.
% package require topotools
% topo readlammpsdata DATA.FILE
You can also visualize the CG configuration with the PSF file (out.psf) and
the CG PDB file (final.pdb):
$ vmd out.psf final.pdb
Please make sure that all the bond connection are correct in the visual ways.