[ESPResSo] Restarting a simulation with clusters of particles

[Lorenzo Isella]

Dear All,
Sorry for coming back indirectly to the topic of clusters of particles.
I typically study a system with 5000 particles (these are normally
called monomers in my field) equally divided into 100 types.
Particles of the same type interact via a strongly binding potential,
but there is no interaction between particles of any two different
types.
If I let a simulation run long enough [I work in a box with periodic
boundary conditions], I will end up 100 clusters each containing 50
particles.
Now, I would like to use these generated clusters as the input of
another Espresso simulation.
I save all the particle positions (and the velocities, but they do not
matter for this discussion here) as returned by Espresso (they will be
outside the box).
However, when I read them, again, Espresso does not understand that
the particles are not far away, but are actually grouped into 100
clusters.
I do not think that asking Espresso to return the folded particle
coordinates and using them for the initial state of the following
simulation is the answer.
For instance [let us make it easy in 1D], right now, if I want to
evaluate the relative distance xr between particles i and j, I have to
use the i and j unfolded coordinates as returned by Espresso and then
calculate xr as in the book by Frenkel & Smit on molecular
simulations, page 68.
Using folded coordinates by itself does not help. Any suggestions here?
Sorry for the long post, Espresso is a wonderful tool for what I have
in mind, but there are a few things I do not grasp yet.
Many thanks

Lorenzo