[gmx-users] Polymers

[James]

Dear Gromacs users,

I am looking to simulate a polymer with a repeating unit cell. Each unit
cell is connected to another unit cell in the same way, with associated
bond and angle restraints, as well as a dihedral force.

Since [bonds], [angles] and [pairs] etc are specified by atom number, it
seems to me that the topology file will become incredibly long if the
polymer is long. I read that one can generate such a system using a
modified .rtp file [1]. If I have understood correctly however, this will
still ultimately result in a very long .top file.

So I would like to ask:
1. Is a long topology file unavoidable? Is specifying by atom name, rather
than number, not possible?
2. Will a long .top file impact performance of mdrun?
3. Is there any limit on the length of the .top file? If not, if anyone has
hit a practical limit (memory, etc), I would be interested to hear about
the experience.

Thank you for your help.
Best regards,
James

[1] http://www.gromacs.org/Documentation/How-tos/Polymers