“GROMACS uses periodic boundary conditions, combined with the *minimum image convention*: only one – the nearest – image of each particle is considered for short-range non-bonded interaction terms. For long-range electrostatic interactions this is not always accurate enough, and GROMACS therefore also incorporates lattice sum methods such as Ewald Sum, PME and PPPM.” (GROMACS Manual: Periodic boundary conditions)

GROMACS Wizard supports the following shapes for space-filling unit cells:

- Cubic,

- Orthorhombic,

- Triclinic,

- Rhombic dodecahedron,

- Truncated octahedron.

You can choose which unit cell to use for the periodic boundary conditions and its size when preparing a system:

The **rhombic dodecahedron** and the **truncated octahedron** are closer to being a sphere than a cube is, and are therefore better suited to the study of an approximately spherical macromolecule in solution since fewer solvent molecules are required to fill the box given a minimum distance between macromolecular images. The rhombic dodecahedron is the smallest and most regular space-filling unit cell. The volume is 71% of the volume of a cube having the same image distance. This saves about 29% of CPU time when simulating a spherical or flexible molecule in a solvent.

Note that GROMACS always keeps the particles in a brick-shaped volume for efficiency. Therefore, when importing the results you might want to choose the proper unit cell representation. When loading GROMACS results, SAMSON will try to automatically detect the type of the unit cell of the system, but you can always modify it in the importer dialog which appears on loading of GROMACS trajectories.

Please refer to GROMACS Manual: Periodic boundary conditions for more information on periodic boundary conditions and unit cell shapes.

## Minimum image convention

Since the periodic boundary conditions are used it is crucial to satisfy the minimum image convention. This means that, for example, a solute should never see its periodic image, otherwise the calculated forces might be wrong. We recommend adding an additional distance of at least 1.0 nm between a solute and the box – that would mean that there is at least 2.0 nm between any two periodic images of solute molecules (please note that molecules in the system, e.g. a protein, might go through various conformations). This distance should be sufficient for just about any cutoff scheme commonly used in simulations but you might need to verify with the force field paper if you are using a custom force field.

Please refer to GROMACS Manual: Cut-off restrictions for more information.

Return to the GROMACS Wizard tutorial.