This has been fixed with Berk's commit c06ee471... The error was not related to GB, but rather a combination of domain decomp.+nm+cut-offs.
Cheers /Per 6 sep 2010 kl. 11.38 skrev Ehud Schreiber: > Dear GROMACS users, > > I am encountering a couple of issues when trying to perform normal mode > analysis in an implicit solvent (GBSA) setting. > I am using version 4.5.1 with double precision; unfortunately I do not have > the single precision version installed for comparison. > > The starting point is a small protein which was energy minimized in two > stages, also in double precision and with GBSA, producing “em2” files. The > mdp file used for the normal mode analysis is the following: > > ----------------- nm.mdp ------------------------- > integrator = nm > nsteps = 1 > implicit_solvent = GBSA > gb_algorithm = Still ; the default > rgbradii = 1.0 ; must be equal to rlist > rlist = 1.0 > coulombtype = cut-off > rcoulomb = 1.0 > vdwtype = cut-off > rvdw = 1.0 > ------------------------------------------------------- > > The command used is > > grompp_d -f nm.mdp -p topol.top -c em2.gro -t em2.trr -o nm.tpr > > which ran fine except for the following note: > > NOTE 1 [file nm.mdp]: > You are using a plain Coulomb cut-off, which might produce artifacts. > You might want to consider using PME electrostatics. > > Then, I tried to run > > mdrun_d -v -nice 0 -deffnm nm > > However, the command seems to be stuck, and the following serious warning is > produced: > > Warning: 1-4 interaction between 64 and 71 at distance 3.114 which is larger > than the 1-4 table size 2.000 nm > These are ignored for the rest of the simulation > This usually means your system is exploding, > if not, you should increase table-extension in your mdp file > or with user tables increase the table size > Maximum force: 3.48709e+06 > Maximum force probably not small enough to ensure that you are in an > energy well. Be aware that negative eigenvalues may occur when the > resulting matrix is diagonalized. > > The em2.gro is definitely not having such a large distance between atoms 64 > and 71: > > 6GLN CB 64 1.879 1.895 2.423 > . > . > 6GLN OE1 71 2.047 1.715 2.642 > > which are only about 0.346 nm apart. Also, the energy minimization, which > used the same options, mutatis mutandis, > > ---------------------- em2.mdp ---------------------------------- > integrator = cg > nsteps = 1000 > nstcgsteep = 40 > implicit_solvent = GBSA > gb_algorithm = Still ; the default > rgbradii = 1.0 ; must be equal to rlist > nstlist = 10 > rlist = 1.0 > coulombtype = cut-off > rcoulomb = 1.0 > vdwtype = cut-off > rvdw = 1.0 > nstenergy = 10 > --------------------------------------------------------------------- > > converged to machine precision having a much smaller maximum force: > > Stepsize too small, or no change in energy. > Converged to machine precision, > but not to the requested precision Fmax < 10 > > Polak-Ribiere Conjugate Gradients converged to machine precision in 0 steps, > but did not reach the requested Fmax < 10. > Potential Energy = -2.71868395521758e+04 > Maximum force = 4.63436548427712e+02 on atom 584 > Norm of force = 1.25296847735530e+02 > > The other problem arises when I try to follow the grompp_d note above and > change in nm.mdp to > > coulombtype = pme > > I then have a fatal error: > > ERROR 1 [file nm.mdp]: > With GBSA, coulombtype must be equal to Cut-off > > Am I doing something wrong or are there still some problems in the new GBSA > option of version 4.5.1? > > Thanks, > Ehud Schreiber. > > -- > gmx-users mailing list gmx-users@gromacs.org > http://lists.gromacs.org/mailman/listinfo/gmx-users > Please search the archive at http://www.gromacs.org/search before posting! > Please don't post (un)subscribe requests to the list. Use the > www interface or send it to gmx-users-requ...@gromacs.org. > Can't post? Read http://www.gromacs.org/mailing_lists/users.php
-- gmx-users mailing list gmx-users@gromacs.org http://lists.gromacs.org/mailman/listinfo/gmx-users Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting! Please don't post (un)subscribe requests to the list. Use the www interface or send it to gmx-users-requ...@gromacs.org. Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
