Faq If you cannot find an answer to your question, let us know

  • What systems is MOSAICS available on?

    Stand-alone MOSAICS is available for Linux and OsX. Source is available so you can compile it yourself for your system. PymoSAICS (GUI) *should* work on Linux, OsX and Windows, see tutorial video of how to install it.

  • What type of simulations/modeling can I run using MOSAICS ?

    When using natural move Monte Carlo, we recommend to perform simulations on nucleic acids at all-atom representation or proteins at course-grained representation as these are the model resolutions that are compatible with our (hierarchical) natural move Monte Carlo protocol. For proteins in atomistic representation we recommend using cartesian degrees of freedom and only simulations aimed to generate local minimum structures using either direct or stochastic methods available in MOSAICS.

  • Can I have overlapping regions in MOSAICS?

    Yes, you define moves in a hierarchichal way, thus you can nest regions.

  • Do all options availibale in MOSAICS is available in PyMOSAICS?

    PyMOSAICS has only a handful number of options so that you can define the regions in Pymol and you may also run a short natural move Monte Carlo simulation to test your choice. Therefore, using PyMOSAICS saves you the time of manually defining the regions. Once the region definitions are available, it is recommended to run production simulations from the command line and not inside the GUI.

  • How long should my naturla move Monte Carlo simulation be?

    When you are just starting off with a structure, we suggest a small number of steps, order of magnitude thousands of MC steps. In this way you will see if your simulation behaves as expected. For example, if you are running parallel tempering Monte Carlo you may check that you have a proper acceptance rate of your individual replicas as well as acceptance rate for replica exchange. If this is not the case you may adjust, natural 'move size' or your temperature ladder. When running production simulations, it depends on the structural observable of interest and you usually run the simulation until the distribution over desired observabled are converged.

  • Structure I loaded fails.

    If you loaded a structure straight from the PDB that is what will happen. Structures need to be processed for MOSAICS and the automation of this process is still a work in progress... In general, MOSAICS expects a .pdb file with all the atoms in the nucleotides (or residues) present and named according to the topology file you are using. Therefore before running a simulation you may check the structure of the topology file chosen for your simulation. For example, atomic models are named according to the UPAC convension. The "Scripts Databse" that you may download with the command line version includes some scripts that can help you. Furthermore, you may be able to: 1) load the structure into the Swiss PDB Viewer -- this fill in missing atoms. 2) Modify terminal ends of EACH CHAIN to make all residues of a given type the same (just like in the system when terminal ends are removed). 3) Use Pymol version 1.7 or higher to allow PyMOSAICS to process the structure. You can also drop us a line if all else fails or use an example structure to learn how they are processed so that you will be able to write your own scripts.