On Tue, Jun 21, 2016, 163邮箱 wrote:
>
> Thank you for your help! I have successfully established the
> corresponding prmtop and inpcrd files of heme-contained protein
> with the heme parameters provided by David A Case. After I had run
> MD simulation, I tried to calculate free energy using mmpbsa.pl
> or mmpbsa.py. Unfortunately, a problem arises which said“Bad atom
> type: FE”when I used mmpbsa.pl and mmpbsa.py modules. I wanted to know
> what should I do to solve this problem and whether there is existing Fe
> parameter when using those two modules.
The mmpbsa programs create a superstructure that attempts to organize
a large number of calculations and automate the analysis prospects. They
can work well for standard situations, but will fail if any one program
in the workflow fails. No one (including me!) has made a full effort (and it
would be a big one) to establish a good framework for finding errors and
reporting them in an understandable way to users, with suggestions for how
to avoid them.
My suggestions:
1. As the manual suggests, run your initial simulations "by hand": create
prmtop files for the complex, receptor and ligand; compute energies for
selected snapshots for each of these three species, and subtract the
receptor and ligand values from that for the complex. Since you are
running each step individually, any problems will probably be much easier
to isolate and fix. Taking the time to look carefully and critically at
the details of the energies and their fluctuations is just as important
and taking the time to carefully and critically examine your simulations
in a visualization program.
2. Try to avoid running GB (which doesn't have any optimized parameters for
metals); choose the options in PB to use the radii in the prmtop files.
3. If all else fails, then do a Google search on "Amber bad atom type", and
you will find many good hints on how to proceed. [Unlike the case with
crosswords, using an internet search engine is not "cheating" when trying to
get your programs to work well.]
....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jun 21 2016 - 05:30:02 PDT
