It strongly depends on your question how you are going to analyze a simulation trajectory.

• Click Options > Macro & Movie > Set target and choose MyDir/MyStructure (no extension!) as the target.

• Click Options > Macro & Movie > Play and choose an analysis macro, e.g. md_analyze.mcr.

This macro will create a self-explanatory table MyDir/MyStructure_Analysis.tab which you can easily import in your favorite data visualization program (Excel, OpenOffice, XMGrace etc.). If you want more than the standard indicators (force field energies and RMSD from the starting structure), open yasara/mcr/md_analyze.mcr in a text editor and create your own version. Note that the energies in the table depend on the force field and cutoff that are chosen in the macro, so you may have to adapt it for your own simulations.

Here are a few hints for the analysis:

• If you obtain an unexpected result, check how reproducible it is by running the simulation a second time with a different random number seed.

• Care has been taken that molecular dynamics trajectories are entirely reproducible, i.e. if you run the same md_run.mcr a second time, using the same YASARA version on the same computer and assigning the same number of processors, you will obtain exactly the same trajectory again. Note however that the reproducibility may get lost when comparing different CPUs, operating systems or YASARA versions: AMD and Intel CPUs support different instruction sets for high performance calculations, which yield slightly different results. Also today's operating systems provide different mathematics libaries, which cause additional deviations. Finally, changes in the YASARA source code may also have a small influence on the results. In short: force field energies calculated for the same scene on different computers are likely to differ in the least significant digits, which are of no relevance anyway. The same is true for the calculated forces, which has a serious impact on simulations however: chaos theory requires that small initial differences grow exponentially with each simulation step, leading to very different trajectories and energies, just as if the simulation had been started with a different random number seed.

• If you compare simulations run at different pH values and obtain quite different results, compare the covalent connectivity first: use the CompareAtom command to find those residues where YASARA assigned differing protonation states during the neutralization experiment, and investigate the region around these residues for an explanation of the structural deviations. Also check the significance of the result by running the simulation a second time with a different random number seed.