On Thu, Apr 16, 2015 at 12:26 AM, Abhishek TYAGI <atyagiaa.connect.ust.hk>
wrote:
> Dear Experts,
>
>
> Installed MMPBSA.py.MPI on gpu cluster successfully, have following
> questions:
>
>
> 1. Running on serial MMPBSA.py I am able to run the mmpbsa.in file with
> all the desired result.
>
> 2. I am able to perform GB and Decomposition calculations without any
> problem using
>
>
> mpirun -np 60 $AMBERHOME/bin/MMPBSA.py.MPI -O -i mmpbsa.in -o
> Final_rresult.dat -do Final_decom.dat -sp solvated.prmtop -cp
> complex.prmtop -rp receptor.prmtop -lp ligand.prmtop -y last5ns.dcd >
> out.log
>
>
> 2. When I run entropy calculations, the problem appears as in form of
> error "Error with quasi-harmonic MMPBSA.py calculations"
>
>
> input file:
>
> Input file for running PB and GB
>
> &general
>
>
> startframe=1, endframe=200, verbose=2, keep_files=2,
> strip_mask=":WAT:SOD:CLA", interval=1, entropy=1,
>
> /
>
> &gb
>
> igb=1, saltcon=0.100,
>
> /
>
> &decomp
>
> idecomp=1,
>
> dec_verbose=1,
>
> /
>
>
> I will be very thankful for the reply.
>
​Look at the _MMPBSA_complex_entropy.out file or any of the files with
"cpptrajentropy" in the name to look for a more complete error message.
Also note that quasi-harmonic analysis will typically require a LOT more
than 200 frames. It effectively does PCA and takes the mass-weighted
principal components as normal modes. And for PCA, you need at *least* 3x
more snapshots as you have atoms just to get enough eigenmodes. And you
typically need a lot more to converge your entropy.
I suggest reviewing literature on the quasi-harmonic approximation before
trying to use it (or interpret results coming from it).
HTH,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Apr 16 2015 - 07:00:03 PDT
