On Fri, 2012-03-30 at 21:37 +0900, 板凱哲彦 wrote:
> Dear AMBER users and developers,
>
> I performed sander for a protein in the free state, which
> was a successful work.
>
> But, for the complex with ligands, I detected the following
> error message;
>
> ----------------------------------------
>
>
> vlimit exceeded for step 0; vmax = 29909.5926
This is a very large velocity. Have you visualized your trajectory to
look for any steric clashes?
Since it happens before the very first step, have you checked your
restart file? Does it have a velocity this large?
> partition error in shake on processor 0
> this processor has atoms 1 through 9100
> atom 9098 is within this range
> atom 9101 is not within this range !
>
>
> ------------------------------------------
>
> In the PDB-formatted file, there is a TER line between the
> 9100 and 9101 atoms, as follows, and it may lead the above
> error, I think.
>
> --------------------------------------------
>
>
> ATOM 9098 C15 L35 579 49.112 52.954 68.279
> ATOM 9099 H7 L35 579 48.964 51.593 69.965
> ATOM 9100 H8 L35 579 50.040 51.978 70.486
> TER
> ATOM 9101 H9 L35 579 48.089 50.908 70.547
> ATOM 9102 C16 L35 579 32.519 59.825 33.305
What is the TER card doing here? Is the ligand in 2 pieces? Are any
atoms greater than 9100 bonded to any atoms before 9100? If so, you
should remove the TER card so leap won't think they are separate
molecules.
It's important for sander that for all SHAKEn bonds, both atoms belong
to the same processor.
HTH,
Jason
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Candidate
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Mar 30 2012 - 07:30:05 PDT
