Question about atomistic and coarse-grained simulation

We are interested in the simulations of soft-matters such as liquid
crystals and colloids.
We think that discussion based on both atomistic and coarse-grained
pictures would be required in order to interpret these phenomena.

We can find a sample simulation of a micelle system in the directory
"<LAMMPS>/examples/micelle/".
This seems to be a coarse-grained simulation of the two-dimension
micelle system.
What kind of molecular system is assumed for this model?

Moreover, please tell me the reverse mapping procedure from this
coarse-grained trajectories to the real atomistic trajectories, if any.

Best regards,
WATANABE

We are interested in the simulations of soft-matters such as liquid
crystals and colloids.
We think that discussion based on both atomistic and coarse-grained
pictures would be required in order to interpret these phenomena.

We can find a sample simulation of a micelle system in the directory
"<LAMMPS>/examples/micelle/".
This seems to be a coarse-grained simulation of the two-dimension
micelle system.
What kind of molecular system is assumed for this model?

nothing specific. it is an extremely simplified picture of amphiphilic
molecules.
you model hydrophilic parts by using a regular lennard-jones and
hydrophobic parts with a lennard-jones with the cutoff in the minimum,
so they are purely repulsive. as you can see, that gets the basic
properties right in a surprisingly convincing fashion.

there are quite a few different models for coarse graining, most of
which are supported by LAMMPS. you may want to check out the relevant
literature. one place to start could be here:
http://dx.doi.org/10.1126/science.1157834

Moreover, please tell me the reverse mapping procedure from this
coarse-grained trajectories to the real atomistic trajectories, if any.

this is model specific and since you lose information when going from
all-atom to coarse grain, you cannot get all that information back.
there should be some papers out there by steve o. nielsen and preston
b. moore on that subject for a specific coarse grain model (the one in
the USER-CG-CMM package or LAMMPS). there is also work of the Schulten
group on that with NAMD using and older version of marrink's martini
model (which is probably the most popular coarse grain model in the
"bio" community).

axel.