Anomalous values of the first six normal modes

We had some more conversation over mail. To summarize:

  1. The eigenvalues of the normal modes not being zero comes likely from the molecule (protein) being large and numerical inaccuracies arising.
  2. To visualize what a normal mode represent it might be useful to make an animation from the eigenvector (see e.g. on this page, using e.g. this VMD tool)