[gmx-users] pdb2gmx disulfide bond in dimer

[s.varriale]

Dear all,
I try to perform a MD simulation of a covalent dimer with gromacs 4.5.4.
I do:
pdb2gmx -f prot.pdb -chainsep ter
and the programme generates a .gro file with disulfide bond.
but when i want to minimize the system, there is an error:

There were 2 inconsistent shifts. Check your topology
Warning: 1-4 interaction between 45 and 387 at distance 8.995 which is 
larger than the 1-4 table size 2.200 nm

if I try to simulate the 2 chains separatly, there is any error.
What can i do?
thanks
Sonia