SIMONAPreprocessor
Gromacs based Monte Carlo - First Steps

How to create your first molecule simulation based on GROMACS.

This is just an example to get to know how this section works and what it does roughly. For more detailed information about all functions of POEM++ see here.
Before you try to do simulations, be sure you know how to use POEM++ or Generic Monte Carlo. Here, only steps that are different will be described.

  • Preparation with GROMACS You need to prepare your system with GROMACS to be able to do Monte-Carlo-Simulations of this system using SIMONA. Please be aware that for MC simulations, only implicit description of solvent molecules is possible as explicit solvent would most likely result in clashes.
    SIMONA reads the structure of the system from a pdb (or mol2) file. Furthermore, it needs the topology (written by GROMACS per default to file topol.top). In the topology file, the forcefield is referenced and these referenced data is read by SIMONA and used for the simulation. You can choose whatever forcefield you like but it can happen that some features of the forcefields are not (yet) implemented in SIMONA and cannot be used since.
    IMPORTANT: you HAVE to make sure that atom naming in the structure and topology file match. If not, you will get errors when the preprocessor tries to sort out all data and gets confused. If you use pdb2gmx for preparation, use its output file (see below).
  • Get and prepare a molecule
    If you have a strucure SIMONA can read and a topology file, you can skip this.
    Get the strucure of your system somewhere (e.g. from the protein data base), make it a pdb file (if it's not) and clean it. Maybe you need to rebuild missing fragments or add missing atoms and maybe you need to delete alternate conformations and models. In the end, you should have a nice pdb that you can process with pdb2gmx. For help on this step, look in the GROMACS manual, read some GROMACS tutorials or the mailing list.
    Process your molecule using pdb2gmx. Make sure to use the correct input file, not to use water and to write output to a .pdb file also (and not to a .gro file as pdb2gmx usually does). Assuming you process 1L2Y using amber99 forcefield, the command may look like this:
    1 pdb2gmx -f 1L2Y.pdb -o 1L2Y_gromacs.pdb -water none -ff amber99
    Possibly, you need to add -ignh to let GROMACS decide where to put hydrogens.
    When pdb2gmx succeeds, you should get (apart from others) the new structure (with correct naming) in file 1L2Y_gromacs.pdb and a topology file called topol.top . These two are needed for the GROMACS import module of the SIMONA preprocessor.
  • Startwindow
    Now start the SIMONA Preprocessor by typing
    1 python YOUR_SIMONA_PATH/python/simonagui.py
    into the shell. If you are still in the folder where you prepared your system, you have to replace YOUR_SIMONA_PATH with the (absolute or relative) path of your SIMONA version.
    In opening window you can choose of which type your simulation should be. Select GROMACS based MC.
  • Input files
    To read in the structure and topology of the molecule, select Configuration at the menu on the left hand side. Type the path of the input-file for the structure in the box beside Input configuration. You may also browse the path by clicking the ... button beside the box. If you have chosen a strucure file, SIMONA will automatically search for a .top file with the same basename or a file called topol.top for the topology. You may choose a different topology file by entering or browsing it using the box besides Input topology. Also type the output-path in the box beside Output xml. To receive a file readable by SIMONA the file extension has to be *.xml.
  • Simulation Settings
    Apart from the step number and the seed, you can specify if you want to use moves like in POEM, if you want to specify moves by hand or if you want to use all dihedral angles defined by GROMACS as moves. The latter one is not recommended, as it may result in destruction of the system but may be useful if you want to edit the output xml file by hand. If you want to do a simulation of protein or DNA (i.e. system that can be simulated using the POEM module), it is recommended to generate moves using POEM. A semi-automated choice of moves like "use all \form#0 angles" is planned for the future but not implemented yet.
  • Forcefields
    For compliance with GROMACS results, you should use the same forcefield as GROMACS does. That means, you should not use any PFF forcefield as these were parametrized differently.
    Usually, this means you should use the following terms:
    1. AMBER-6,12-Lennard-Jones
    2. AMBER-6,12-Lennard-Jones for 1,4-neighbours
    3. AMBER Coulomb electrostatics (with exclude list)
    4. AMBER Coulomb electrostatics for 1,4-neighbours
    If you want to use the implicit solvent model, add these terms:
    1. Born radii
    2. PFF - PowerSasa
    3. ALPB Still
    4. Non-polar solvation energy
  • Further steps and results
    The remaining steps are fairly similar to those in POEM. Choose moves as you want, specify the output, expert options (if you need those), run the simulation use poempp and analyse it with the tool of your choice.