On Mon, Oct 05, 2015, Jérémie Knoops wrote:
>
> Otherwise, what is the easiest method to create parameters and
> topology files for MD simulations in explicit solvent (DNA, ligand and
> DNA+ligand) followed by a MM-GBSA analysis (separate trajectory
> approach)?
> I don't know how to create them easily without using the combine command.
I personally never use the combine command, and I suspect that other Amber
developers are in the same boat, which is why it is certainly possible that
there are cases where it doesn't work the way it should.
The recommended (by me!) way:
1. Create a pdb file with the entire system you want; use TER cards to separate
chains if there is any abiguity.
2. Load various library and frcmod files that will be needed; then use a
single "loadPdb" command to create the entire system.
3. If you want smaller pieces (e.g. for a multiple trajectory energy
analysis), just modify the input pdb file to have the parts you want.
This avoids any use of copy and combine commands. It also seems (to me,
anyway) to be much simpler than the procedure you propose:
> >
> > DNA = sequence { DT5 DT DT3 }
> > DNA2 = copy DNA
> > saveoff DNA DNA.lib
> > saveamberparm DNA DNA.prmtop DNA.inpcrd
> > saveoff DNA2 DNA2.lib
> > saveamberparm DNA2 DNA2.prmtop DNA2.inpcrd
None of the above commands should be needed.
> >
> > LIG = loadmol2 Mol-sm_m9-c1.mol2
> > DNA-LIG= combine {DNA LIG}
> > saveoff DNA-LIG DNA-LIG.lib
> > saveamberparm DNA-LIG DNA-LIG.prmtop DNA-LIG.inpcrd
> > DNA-LIG2= combine {DNA2 LIG}
> > saveoff DNA-LIG2 DNA-LIG2.lib
> > saveamberparm DNA-LIG2 DNA-LIG2.prmtop DNA-LIG2.inpcrd
> >
I'm guessing that something odd is happening in the copy step; if you want
to explore this, just use the following definition of DNA2, in place of what
you have above:
DNA2 = sequence { DT5 DT DT3 }
See if that gives different behavior.
As I said in my earlier email, this is arguably a bug that should be examined.
In the meantime, please see if either of the two workarounds suggested above
will help.
....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Oct 05 2015 - 14:30:03 PDT
