Hi there!
Ok, I've continued to try, and I think I've got a solution. I'd still really appreciate any input you might have, though, so thank you in advance!
Basically, as you can see from my previous message, all restraints are centered around a single residue in one of the domains. As a result, that residue was subject to massive energies (at least that was my assumption). I therefore restrained the region around that residue (7 restrained amino-acids, 3 C-terminal, 3 N-terminal) with a 5 KCal/molA positional restraint.
After I run a simulation under those conditions, most of the movement now occurred on the other domain, as expected. The problem was that now the other domain was being torn apart by the NMR restraints. I decided to introduce a restraint weight gradient, to keep that domain happy. I tested several gradients, but it looks like a gradient over 2 ns (dt=0.002) from 0.1 to 1 does the trick.
I hope this might help someone else in designing a similar system, and I hope that, if I've made any mistakes in my assumptions/procedure, you can point them out for me!
Thanks a lot,
Manuel
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Jan 08 2015 - 17:00:02 PST
