[lammps-users] help: about creating PSF file

Dear all,
I have a problem with creating data file from a PDB.
I have used successfully ch2lmp tool for creating data file from PDB based on CHARMM forcefield for proteins and DNAs.
There is a step in converting with ch2lmp (creating PSF file) which I have used psf generator of VMD to do that step.
But now I want to create a data file from a PDB of another polymer with this forcefield and unfortunately I couldn’t used VMD (I think because of missing suitable input file).
I want to know if there is somebody who worked with this tool? and can anybody help me to know how to create a PSF file from a PDB with VMD?
Also in ch2lmp tool there is a short instruction to create psf file manually. can anybody help me to find a more detailed report about it?

Thanks in advance and with the best regards.

Arman

Dear all,
I have a problem with creating data file from a PDB.
I have used successfully ch2lmp tool for creating data file from PDB based
on CHARMM forcefield for proteins and DNAs.
There is a step in converting with ch2lmp (creating PSF file) which I have
used psf generator of VMD to do that step.

But now I want to create a data file from a PDB of another polymer with this
forcefield and unfortunately I couldn't used VMD (I think because of missing
suitable input file).
I want to know if there is somebody who worked with this tool? and can

the authors of ch2lmp are listed in the README file.

anybody help me to know how to create a PSF file from a PDB with VMD?

there are tutorials for all kinds of VMD functionality on the VMD homepage.
the problem is not only to have just a .psf file (that can always be done
with a simple command), you first have to provide all the correct information
so that the resulting .psf file contains all the information that you need to
run the simulation successfully. at this stage, it would probably be easier
to use the topotools plugin in VMD to write out a lammps data file directly.

however, there is not really fully automatic way to do this. the
information that you need to to perform a simulation is much
more than what a .pdb file contains. so you have to somehow
enter this. either manually, or with some smart scripting.

if you don't have the input to use VMD's psfgen, you won't have
the required input to run ch2lmp as well, as both use the same
sources to determine the missing information.

Also in ch2lmp tool there is a short instruction to create psf file
manually. can anybody help me to find a more detailed report about it?

i don't think that such a thing exists. but the process is fairly
straightforward.
the biggest problem is generally to match atom names in your input
with the corresponding atom types in the data file. for non-conventional
input, you'll have to do the assignments essentially manually or write
a script that does it for you.

axel.