[gmx-users] solvation problem

[Marc Bruning]

hello,

while doing an energy minimization of a protein (with crystal waters) i
get the following error message:

"Fatal error: number of coordinates in coordinate file (epsp_b4sim.gro,
49069)
             does not match topology (epsp.top, 49072)"

further up i find the following (during solvation):

"Added 14837 molecules
Generated solvent containing 44511 atoms in 14837 residues
Writing generated configuration to epsp_b4sim.gro

Back Off! I just backed up epsp_b4sim.gro to #epsp_b4sim.gro#
Protein in water t= 254.00000   + generated by addprodrg.pl from
epsp_model_start.gro and  s3p.gro    + generated by addprodrg.pl from
epsp_model_s3p.gro and  glp.gro    + generated by addprodrg.pl from
epsp_model_glp.gro and  hoh.gro

Output configuration contains 49069 atoms in 15262 residues
WARNING: masses will be determined based on residue and atom names,
         this can deviate from the real mass of the atom type
Opening library file /usr/local/gromacs/share/top/atommass.dat
Volume                 :     507.828 (nm^3)
Density                :     1031.39 (g/l)
Number of SOL molecules:  14838 "

it seems to me that the program firts adds 14837 water molecules but in
the end defines the number of SOL-molecules as 14838! while you can find
coordinates for 14837 waters in the gro-file, in the top-file the number
of solvent is defined as 14838. that one water molecule (with 3
coordinates) difference causes in my opinion the above stated error. the
change of the number of water molecules happens during one run with
genbox. there are no additional steps in between. maybe it happens
because i include crystal waters and the program somehow mixes the two
kinds of waters up? to not loose my crystal waters in the rest of the
solvent i use an own itp-file for it adapted from the flexspc.itp. maybe
thats the  source of the error?

thanks in advance for any help,
regards
marc bruning



--
marc bruning
mpg-asmb