From: Wang, Boyang (bwang9_at_uic.edu)
Date: Wed Jan 24 2007 - 12:59:54 CST
Dear all,
There is a problem of mine that I have to ask you for help with.
Dr. Gengbin Zheng was very nice and helped us last year to install a NAMD
from "cvs" in our local cluster running in parallel processors. This
version of NAMD can recognize CHARMM22 force field and run MD simulations
in multiple processors.
Now we are going to run nucleic acid dynamics, and we are using CHARMM31
force field. The NAMD running in parallel (using MPI) that Gengbin
installed in our cluster does not seem to recognize the CHARMM31 force
field, with the error as follows,
Info: PARAMETER file: CHARMM format!
Info: PARAMETERS par_all27_prot_na.prm
Info: BINARY COORDINATES r3.restart.coor
FATAL ERROR: BAD IMPROPER FORMAT IN CHARMM PARAMETER FILE
LINE=*CMAP*
------------- Processor 0 Exiting: Called CmiAbort ------------
Reason: FATAL ERROR: BAD IMPROPER FORMAT IN CHARMM PARAMETER FILE
LINE=*CMAP*
The NAMD 2.6b1 binary running on single processor does not recognize
CHARMM31 force field, with the same error in the output as shown above.
When we used the newest NAMD 2.6 binary, NAMD 2.6 is running well with
CHARMM31 force field in a single processor.
I asked Gengbin if it is possible to install the newest version of NAMD
running in parallel in our cluster so that we can run nucleic acid
dynamics in 4 processors. However, Gengbin believed that NAMD from "CVS"
is the newest version that is newer than what is released. I don't know
what is "CVS" and I wonder what is required to run NAMD in parallel for
CHARMM31 force field.
Could you give me a little hint so that I can learn something more from
this situation?
Thank you very much for your time and considerations!
Sincerely,
Boyang
Univ Illinois,
Chicago.
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:44:19 CST