psf-pdb-builder-python

Python script creating PSF and PDB files for running molecular dynamics simulations in NAMD, especially with usage of Drude particles.

Requirements

Only standard library for Python 3.6 has been used, there are no further requirements.

Running

python3 main.py path_to_packmol_file [parameters]

Available parameters:

• -psf file_name - produce PSF file with a given name
• -pdb file_name - produce PDB file with a given name
• -t or --tinker - for every residue .conn files are also provided
• -d or --drude - include declared Drude particles

Example running commands are presented in run_examples.sh.

Input files

To correctly run the program there are some input files needed:

• Packmol format input file which name is specified when running the program
• .xyz file with atoms position for the whole system
• .dat and .xyz files describing every residue of the system
• .conn files for every residue if -t option is set on

Packmol input

Example:

seed 1030
tolerance 2.0
output li-ec-01.xyz
filetype xyz
structure li.xyz
  number 4
  inside cube -22.0 -22.0 -22.0 44.0
end structure
structure ec.xyz
  number 46
  inside cube -22.0 -22.0 -22.0 44.0
end structure
structure tfsi.xyz
  number 4
  inside cube -22.0 -22.0 -22.0 44.0
end structure

Only lines beginning with output, structure and number are parsed.

• output - specifies name of the .xyz file containing atom positions in the whole system
• structure - for every residue of the system a section beginning with structure and ending with end should be specified. In the line with structure keyword name of the .xyz file containing atom positions of given residue should be given. Additionally number gives the number of repetitions of a given residue in the whole system

The order of structure declarations should be the same as they order in the .xyz file!

.dat input file

For every residue declared as structure in the Packmol input file .dat file with the same name as specified .xyz file should be present. Example:

EC
O3 -0.53 15.9994
Ccc -0.05 12.0107
O3 -0.53 15.9994
C8 -0.05 12.0107
C8 -0.05 12.0107
O2 -0.53 15.9994
H1 0.13 1.00794
H1 0.13 1.00794
H1 0.13 1.00794
H1 0.13 1.00794
BONDS:
1 2
1 5
2 3
2 6
3 4
4 8
4 9
4 5
5 7
5 10
END

File structure description:

• first line - residue name (max 3 characters)
• for every atom one line should be specified in format atom_symbol_for_NAMD charge mass in the same order as they appear in the .xyz file
• at the end of the file BONDS section could be specified containing in separate lines indices of atoms forming bonds (the first atom has an index equal to 1), this section must be ended with END keyword
• if BONDS section is not specified the program will try to detect bonds automatically (if atoms are closer than 1.5 angstroms) or if program is launched with -t option bond information will be read from .conn files

To declare Drude atoms one needs to give d symbol and polarizability at the end of the line of the atom which should have the Drude atom, for example:

FSI
Sf 1.02 32.065 d 1.00
Ff -0.13 18.9984 d 1.00
Of -0.53 15.9994 d 1.00
Of -0.53 15.9994 d 1.00
Nf -0.66 14.0067
Sf 1.02 32.065 d 1.00
Ff -0.13 18.9984 d 1.00
Of -0.53 15.9994 d 1.00
Of -0.53 15.9994 d 1.00

In addition, when running program for given file but without specifying -d option, information about Drude particles will be skipped.

.xyz file

For every residue declared in Packmol file a .xyz file containing list of atoms in it should be specified in standard .xyz format, example:

10

  O         -6.311309       -5.532314       -0.465320       
  C         -7.260238       -5.417400        0.432049       
  O         -7.345047       -4.170005        0.997326       
  C         -6.302795       -3.379001        0.350445       
  C         -5.446716       -4.423542       -0.336933       
  O         -7.870834       -6.396057        0.818066       
  H         -5.065063       -4.206253       -1.341663       
  H         -6.751815       -2.767502       -0.441364       
  H         -5.703414       -2.825454        1.136486       
  H         -4.593200       -4.729328        0.253094       

• first line - atoms number
• second line - for comments, skipped during parsing
• every next line - atom symbol and its Cartesian coordinates

When -t option was set on, the .xyz files are expected to be in Tinker format which differs a bit from that presented above, it does not contain a separate commentary line and also every atom line begins with atom index, example:

           9  molden generated tinker .xyz (mm3 param.)
 1  S     1.401416    0.159073   -0.087487     66  2  3  4  5
 2  F     1.714650   -1.415751    0.421888     64  1
 3  O     1.371717    0.926102    1.174689     61  1
 4  O     2.468252    0.411380   -1.076466     61  1
 5  N    -0.000023    0.000005   -0.909775     59  1  6
 6  S    -1.401385   -0.159093   -0.087474     66  5  7  8  9
 7  F    -1.715104    1.415771    0.421979     64  6
 8  O    -1.371117   -0.926126    1.174700     61  6
 9  O    -2.468382   -0.411342   -1.076298     61  6

Tinker file (.conn)

If -t option was set on, for every residue there should exist a .conn file (with the same name as the .xyz file for this residue) in the format of an output of Tinker analyze module, example is provided in this file. From this file information about bonds, angles and torsional angles is read.

Examples description

Example input files are located in subdirectories in tests, commands used to launch the program for them are specified in run_examples.sh file

• li-ec - basic example, without Drude particles and without Tinker output files with bonds specified in .dat files
• fsi-tinker - example of system with Tinker output file
• fsi-tinker-drude - the same as the above, but with additionally declared Drude particles