Hi,

This link should cover most of the programs out there if you want to know how they are licensed:

http://en.wikipedia.org/wiki/List_of_quantum_chemistry_and_solid-state_physics_software

Cheers

As I have the force field of the molecule and also its pdb structure file, I can find the potential energy of a system of two molecules by MD simulation. I don't know whether there is any other simpler method. Thank you Jan for the link. Can you be more specific ? The list contains more than 70 softwares !

Hi, I would suggest you use Firefly (http://classic.chem.msu.su/gran/gamess/index.html), a free quantum chemistry software, to calculate the interactions between two molecules. My experience of Firefly is that it is quite fast and reliable for a free software. The way to do this would be to perform a excited-state calculation and look at the splitting of the transition energies - half of this value is the molecular interaction energy. I am not sure that MD would give a very accurate results, and I am not sure you could easily control the separation and the molecular geometries with MD, which is why I suggest a quantum chemistry approach instead. I have published an article with this kind of approach if you want some more details (it's on my profile).

so basically you want to calculate PMF (potential of mean force), right?

If you consider a molecule to be a single particle. Then what I want is to calculate the pair potential function of two interacting particles (i.e. molecules).

it means you need effective pair potential which you can calculate using the potential of mean force. Using MD packages like DL_POLY or LAMMPS you can easily get the PMF but you need to understand the logic that what are you doing. In LAMMPS, it can be done by number of ways (Steered Molecular Dynamics or you can manually bring two molecules at certain distance by picking some fix in LAMMPS and calculate energy). DL_POLY v4.04 gives PMF directly.

You can do this in two ways: by quantum mechanics (QM) and molecular dynamics (MD). In QM, locate in the software manual (eg. firefly, ORCA), how to make scans. In MD, you need to check if the parameters necessary for the molecules of interest exists, to apply the procedure suggested by Gourav Shrivastav

Orca is free, robust and available on different platforms. http://cec.mpg.de/forum/

Look at its manual for scan.

The CamCASP program, which is based on SAPT(DFT), will probably do what you want. See http://www-stone.ch.cam.ac.uk/programs/camcasp.html.

This question goes right to the heart of one of the major problems in molecular cluster simulation - SOLVATION. Our current ways of simulating solvation are weak, and almost ridiculous. The "ether" method is just a joke.

As we see from the replies, there is no simple solution to the problem short of actually making the cluster, in silico, and then then doing the measurements while varying the geometry of the cluster.

Looking at the responses, almost everybody assumed the system in the gas phase. Nice observation Vernon.

I'm not really sure I understand what molecular cluster simulation are - I guess that's a type of calculation using Molecular Dynamics (MD). As said above, you have the choice of using either MD or Quantum Chemistry (QC) to calculate molecular interactions. MD usually requires parametrisation of a force-field to get reasonnable answers; this parametrisation is usually done from preliminary QC calculations. That's why in your case I think you should use QC, using MD is too complicated for what you want to achieve. In addition, with QC you can simply include solvent using the PCM method. The influence of solvent on molecular interactions has been studied in details by Curutchet, among others (for instance http://dx.doi.org/10.1021/jp072540p )

Your interaction energy sounds like a single point property (i.e. the quantity you seek can can be computed from a single geometry of the interacting dimer.) To obtain the interaction energy as a function of the separation, you would run a series of such calculations and vary the distance between the molecules however you like (center of mass, closest contact or something else.)

For example, if your system is small (

make a grdf

1) http://manual.gromacs.org/programs/gmx-enemat.html

2) use the Calculate Interaction Energy” protocol encoded in Discovery Studio

Fragment molecular orbitals. For small molecules FMO is quite straightforward, for proteins you gota get FEDOROV software to adapt protein data base (pdb) format to GAMESS format . There you get correlation energy, dispersion energy, dipole moments (as a function of energy), and and also a rational way of calculating energies of fragments and the multifragment complex. Watch FEDOROV HPage and Mark Gordon site.