Re: Stray PME grid charges detected

From: yi wang (yiwang_at_ks.uiuc.edu)
Date: Thu Mar 15 2007 - 18:42:59 CDT

Hi Simon,
       I can't tell what exactly is wrong from your post, however,
I've just experienced a similar problem with a similar error message.
What happened was that my topology file .psf and my coordinate
file .pdb do not match with each other---after I made some changes
with them. Since you said this was the first time you simulated with
a ligand, I suppose there may be some change to the topology file as
well? I suggest you check that part of the file carefully and see if
there is something missing---does your starting point of the system
look normal when loaded in VMD? Are all the bond/angle/dihedral terms
properly defined in the topology file?
      Please let us know how it goes, I'd be interested to know what
was causing the problem.

Best,

yi

On Jan 9, 2007, at 12:06 PM, Simon Watson wrote:

> Hi,
>
> I have seen that several other people on this forum have been
> affected by the same problem; I have followed all the threads, and
> attempted all the suggestions but I am still suffering from the
> same problem.
>
> My system is composed of HIV protease, bound to saquinavir, in a
> water box 81.93 by 90.55 by 81.80. The PME grid size was
> originally 81 by 90 by 81 (x,y,z), and the protein and ligand are
> restrained during the minimisation. When I tried minimising, I
> attained the following output:
>
> ERROR: Stray PME grid charges detected: 0 sending to 7 for planes
> 18 19
> ERROR: Stray PME grid charges detected: 0 sending to 6 for planes
> 16 17
> ERROR: Stray PME grid charges detected: 0 sending to 23 for planes
> 63 64 65
> ERROR: Stray PME grid charges detected: 0 sending to 22 for planes
> 60 61 62
> Warning: Bad global exclusion count, possible error!
> Warning: Increasing cutoff during minimization may avoid this.
> Warning: Bad global bond count!
> Warning: Bad global angle count!
> Warning: Bad global dihedral count!
> Warning: Stray PME grid charges ignored!
> PRESSURE: 122 -1.47246e+07 1.52151e+06 -1.38893e+06 1.5201e+06
> -1.57702e+07 282057 -1.39157e+06 279814 -1.32042e+07
> GPRESSURE: 122 -2.35315e+08 -1.20036e+07 1.84397e+07 -1.2128e+08
> 5.7502e+06 -6.52172e+07 -5.55886e+07 6.14859e+06 -3.7112e+07
> ENERGY: 122 99999999.9999 256946.1353
> 8404.9533 0.0000 -1499602.7597 1171537.4338
> 492703.9185 0.0000 0.0000
> 99999999.9999 0.0000 99999999.9999 99999999.9999
> 0.0000 -14566345.2938 -88892404.2323 606823.5931
> -14566345.2938 -88892404.2323
>
> I have attached my minimisation configuration file. This is the
> first time I have tried simulating with a ligand - all previous
> simulations were apo-proteins. For this reason I have had to set
> up the system in AMBER, though I am simulating with NAMD.
>
> As suggested, I tried increasing the cutoff from 12 to 14, 16, 18,
> but this didn't help. I also tried increasing the size of the PME
> grid, ensuring it stayed a multiple of 2,3 and 5 in each direction,
> but this also didn't help.
>
> I found on the NAMDWiki Troubleshooting page that it could be
> because of an atom (probably hydrogen) moving to the other side of
> the cell, making a bond extend across the cell. I loaded up
> the .pdb and .crd file in VMD and didn't see any very long bonds,
> except for 4 at the N-terminal amino acid that extend out very far
> out of the box, but these are present even in different simulations
> that run with no problems. I also checked for uninitialised
> coordinates, but they all seemed okay.
>
> I am not an experienced user of NAMD, and so was hoping that
> someone could shed some light on any mistakes I have made or
> potential fixes I have overlooked..
>
> Thank you, in advance, for any help.
>
> Simon.
>
> <eq0.conf>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:44:28 CST