On Tue, Sep 11, 2012, Anna Bauß wrote:
>
> I'm trying to do a little TI with Crambin. I want an extra electron to
> transfer from one phenylalanine residue to a thyrosine residue.
> Therefore I modified these residues charges in the mol2 files in such a
> way that the additional electron is located in the phenyl ring.
> The needed frcmod and lib files where done following the tutorials.
It's not very clear what you did here. Molecular mechanics potentials are not
designed to work with radicals. The sorts of results you see should not arise
from small changes in charges, but it's not clear whether or not you made
changes requiring a frcmod file.
>
> The first run I did was interrupted because of SHAKE errors, therefore I
> turned SHAKE off.
Don't do that: find what is causing the SHAKE errors. The usual advice:
Run lambda=0 and 1 calculations in serial mode, with ntpr=1 and nstlim=100.
Make sure the structures are OK. If this all works, make sure you have fully
equilibrated one of the end points to a good temperature and pressure, before
you try TI.
If you have problems, use the "checkoverlap" command in ptraj to make sure
that you have no bad non-bonded contacts.
With TI, use just two threads, again set ntpr=1 and nstlim=100. Look
carefully for where the problems might exist.
....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Sep 11 2012 - 05:30:05 PDT
