2016-08-30 13:18 GMT-06:00 Chang Woon Jang <[email protected]>: > Dear Christoph, > > I think that bond potentials seem to be good with U shape using > csg_call --options options.xml --ia-type bond --ia-name bond > convert_potential gromacs Yeah, the attached bond potentials actually look decent.
I have another idea, see below... > > > The following potentials in figures are obtained from csg_call. I am not > sure of forces though. With all those potentials, now it does not run even > single iteration. > > > > Started mdrun on rank 0 Tue Aug 30 15:10:53 2016 > Step Time Lambda > 0 0.00000 0.00000 > > > Energies (kJ/mol) > Tab. Bonds Tab. Angles Tab. Dih. LJ (SR) Coulomb (SR) > 5.01406e+03 1.41421e+05 2.15701e+02 2.49642e+03 0.00000e+00 > Potential Kinetic En. Total Energy Temperature Pressure (bar) > 1.49147e+05 3.66788e+03 1.52815e+05 3.02256e+02 1.68470e+03 > > DD step 9 load imb.: force 6.7% > > > Not all bonded interactions have been properly assigned to the domain > decomposition cells > > A list of missing interactions: > Tab. Angles of 856 missing 1 This could be a problem too. That means there is a missing interaction for one angle! Christoph > > Molecule type 'TDM' > the first 10 missing interactions, except for exclusions: > Tab. Angles atoms 1 3 2 global 201 203 202 > > ------------------------------------------------------- > Program gmx, VERSION 5.0.8-dev-20151014-1f04b58 > Source code file: > /home/chang/PACKAGE/votca/src/gromacs/src/gromacs/mdlib/domdec_top.c, line: > 394 > > Fatal error: > 1 of the 1720 bonded interactions could not be calculated because some atoms > involved moved further apart than the multi-body cut-off distance (2.02391 > nm) or the two-body cut-off distance (2.02391 nm), see option -rdd, for > pairs and tabulated bonds also see option -ddcheck > For more information and tips for troubleshooting, please check the GROMACS > website at http://www.gromacs.org/Documentation/Errors > ------------------------------------------------------- > > Best regards, > Changwoon Jang > > > > On Tue, Aug 30, 2016 at 2:54 PM, Christoph Junghans <[email protected]> > wrote: >> >> 2016-08-30 12:33 GMT-06:00 Chang Woon Jang <[email protected]>: >> > Dear Christoph, >> > >> > It runs smoothly up to 12500 iterations. After that the error still >> > exist >> > as follows. Luckily, 10K iterations do not produce the error. >> The below error means a bond got stretched more than 2.02391 nm, which >> seems a bit unphysical. >> You can look at your bond distributions if 2.02391 nm has a >> probability greater than 0. >> I guess, you will have to pick the min and max more carefully the xml >> file for the bonds when using csg_call. >> As rule of thumb min and max should lie on a downwards slopping, but >> not quite 0 yet, part of the distribution. >> After all the extrapolation csg_call does (see manual for details) the >> potentials should look like a deformed U or W, you can check that by >> look at the xvg output file. >> >> Christoph >> > >> > >> > >> > DD step 12499 load imb.: force 5.6% >> > >> > Step Time Lambda >> > 12500 6.25000 0.00000 >> > >> > Energies (kJ/mol) >> > Tab. Bonds Tab. Angles Tab. Dih. LJ (SR) Coulomb >> > (SR) >> > 1.21732e+03 8.55168e+02 1.85141e+02 2.66780e+03 >> > 0.00000e+00 >> > Potential Kinetic En. Total Energy Temperature Pressure >> > (bar) >> > 4.92543e+03 3.80913e+03 8.73456e+03 3.13896e+02 >> > 1.21927e+03 >> > >> > >> > Not all bonded interactions have been properly assigned to the domain >> > decomposition cells >> > >> > A list of missing interactions: >> > Tab. Dih. of 108 missing 1 >> > >> > Molecule type 'JMT' >> > the first 10 missing interactions, except for exclusions: >> > Tab. Dih. atoms 1 2 3 4 global 927 928 929 >> > 930 >> > >> > ------------------------------------------------------- >> > Program gmx, VERSION 5.0.8-dev-20151014-1f04b58 >> > Source code file: >> > /home/chang/PACKAGE/votca/src/gromacs/src/gromacs/mdlib/domdec_top.c, >> > line: >> > 394 >> > >> > Fatal error: >> > 1 of the 1720 bonded interactions could not be calculated because some >> > atoms >> > involved moved further apart than the multi-body cut-off distance >> > (2.02391 >> > nm) or the two-body cut-off distance (2.02391 nm), see option -rdd, for >> > pairs and tabulated bonds also see option -ddcheck >> > For more information and tips for troubleshooting, please check the >> > GROMACS >> > website at http://www.gromacs.org/Documentation/Errors >> > ------------------------------------------------------- >> > >> > Best regards, >> > Changwoon Jang >> > >> > >> > On Tue, Aug 30, 2016 at 2:22 PM, Chang Woon Jang >> > <[email protected]> >> > wrote: >> >> >> >> Dear Christoph, >> >> >> >> Did you also change original bond and angle tables to gromacs >> >> format? >> >> >> >> If so, how do I determine the min and max values for bond and angle >> >> tables? >> >> >> >> >> >> >> >> Thank you. >> >> >> >> Best regards, >> >> Changwoon Jang >> >> >> >> On Tue, Aug 30, 2016 at 2:04 PM, Christoph Junghans >> >> <[email protected]> >> >> wrote: >> >>> >> >>> Hi, >> >>> >> >>> thanks for sending the potentials. >> >>> Here is what I did: >> >>> $ ls table_d*.txt >> >>> table_d1.txt table_d2.txt table_d3.txt table_d4.txt >> >>> # These are the potentials out of csg_boltzmann >> >>> $ cat table_d.xml >> >>> <cg> >> >>> <bonded> >> >>> <!-- name of the interaction --> >> >>> <name>dihedral</name> >> >>> <min>-3.141</min> >> >>> <max>3.141</max> >> >>> <step>0.001</step> >> >>> </bonded> >> >>> <inverse> >> >>> <gromacs> >> >>> <pot_max>1e8</pot_max> >> >>> <table_bins>0.001</table_bins> >> >>> </gromacs> >> >>> </inverse> >> >>> </cg> >> >>> $ for i in table_d*.txt; do csg_call --options table_d.xml --ia-type >> >>> dihedral --ia-name dihedral --sloppy-tables convert_potential gromacs >> >>> $i ${i%.txt}.xvg; done >> >>> This basically does: >> >>> csg_call --options table_d.xml --ia-type dihedral --ia-name dihedral >> >>> --sloppy-tables convert_potential gromacs table_d1.txt table_d1.xvg >> >>> csg_call --options table_d.xml --ia-type dihedral --ia-name dihedral >> >>> --sloppy-tables convert_potential gromacs table_d2.txt table_d2.xvg >> >>> csg_call --options table_d.xml --ia-type dihedral --ia-name dihedral >> >>> --sloppy-tables convert_potential gromacs table_d3.txt table_d3.xvg >> >>> csg_call --options table_d.xml --ia-type dihedral --ia-name dihedral >> >>> --sloppy-tables convert_potential gromacs table_d4.txt table_d4.xvg >> >>> Please note you need the "--sloppy-tables" option as you want to >> >>> ignore the third column of table_d1.txt >> >>> Alternatively you could replace the 3rd column by "i": >> >>> $ awk '/^[^#]/{print $1,$2,"i"}' table_d1.txt > fixed_table_d1.txt >> >>> Then a I ran gromacs >> >>> $ gmx mdrun -tableb table_[abd]*.xvg -v >> >>> Please mind that this is gromacs 2016. which is why you need the >> >>> "-tableb" option. (Gromacs 5.0 was made obsolete a while back by its >> >>> developers.) >> >>> >> >>> Seems to work for at least 10k steps. >> >>> >> >>> Cheers, >> >>> >> >>> Christoph >> >>> >> >>> >> >>> >> >>> 2016-08-29 20:52 GMT-06:00 Christoph Junghans <[email protected]>: >> >>> > 2016-08-29 15:19 GMT-06:00 Chang Woon Jang >> >>> > <[email protected]>: >> >>> >> Dear Christoph, >> >>> >> >> >>> >> Can I run ibi without dihedrals? Actually, I do not need >> >>> >> dihedral >> >>> >> potentials even though CG molecules have dihedrals. Or, any >> >>> >> suggestions for >> >>> >> debugging the dihedrals. Actually, the warning signs disappeared >> >>> >> but I >> >>> >> have >> >>> >> still the same Fatal error. >> >>> > Please send me the table and the xml off-list and I will have a >> >>> > look. >> >>> > >> >>> > And of course you can run IBI on a subset of the interactions, see >> >>> > <https://github.com/votca/csg-tutorials/tree/master/hexane/ibi> >> >>> > In this example only the non-bonded interactions are iterated. >> >>> > >> >>> > Christoph >> >>> >> >> >>> >> Thank you. >> >>> >> >> >>> >> Best regards, >> >>> >> Changwoon Jang >> >>> >> >> >>> >> On Mon, Aug 29, 2016 at 4:30 PM, Christoph Junghans >> >>> >> <[email protected]> >> >>> >> wrote: >> >>> >>> >> >>> >>> 2016-08-29 13:32 GMT-06:00 Chang Woon Jang >> >>> >>> <[email protected]>: >> >>> >>> > Dear Votca Users, >> >>> >>> > >> >>> >>> > I have four dihedral distributions directly from >> >>> >>> > csg_boltzmann. >> >>> >>> > However, >> >>> >>> > only one dihedral distribution gives waning sign during IBI run >> >>> >>> > as >> >>> >>> > follows. >> >>> >>> > >> >>> >>> > WARNING: For the 359 non-zero entries for table 0 in >> >>> >>> > table_d1.xvg >> >>> >>> > the >> >>> >>> > forces >> >>> >>> > deviate on average 196% from minus the numerical derivative of >> >>> >>> > the >> >>> >>> > potential >> >>> >>> That means something went wrong in the table conversion! >> >>> >>> > >> >>> >>> > >> >>> >>> > >> >>> >>> > Then, the IBI fails with Fatal error: 1 of the 1720 bonded >> >>> >>> > interactions >> >>> >>> > could not be calculated because some atoms involved moved >> >>> >>> > further >> >>> >>> > apart >> >>> >>> > than >> >>> >>> > the multi-body cut-off distance (2.02391 nm) or the two-body >> >>> >>> > cut-off >> >>> >>> > distance (2.02391 nm), see option -rdd, for pairs and tabulated >> >>> >>> > bonds >> >>> >>> > also >> >>> >>> > see option -ddcheck >> >>> >>> That is most likely due to the above warning. >> >>> >>> > >> >>> >>> > >> >>> >>> > I am not sure why only one dihedral gives this warning and >> >>> >>> > prevent >> >>> >>> > IBI >> >>> >>> > run. >> >>> >>> > Do you have any comments on that? >> >>> >>> IBI for dihedral interactions is a bit tricky and not very well >> >>> >>> testing in VOTCA, so you will need to do a bit a debugging on your >> >>> >>> side. >> >>> >>> >> >>> >>> Christoph >> >>> >>> >> >>> >>> > >> >>> >>> > I have converted dihedral distribution radian to angle (only >> >>> >>> > first >> >>> >>> > column) >> >>> >>> > due to the ibi requirement just using unit convert >> >>> >>> > x*180/3.14 >> >>> >>> > (x is >> >>> >>> > radian). >> >>> >>> > >> >>> >>> > Thank you. >> >>> >>> > >> >>> >>> > Best regards, >> >>> >>> > Changwoon Jang, >> >>> >>> > >> >>> >>> > -- >> >>> >>> > You received this message because you are subscribed to the >> >>> >>> > Google >> >>> >>> > Groups >> >>> >>> > "votca" group. >> >>> >>> > To unsubscribe from this group and stop receiving emails from >> >>> >>> > it, >> >>> >>> > send >> >>> >>> > an >> >>> >>> > email to [email protected]. >> >>> >>> > To post to this group, send email to [email protected]. >> >>> >>> > Visit this group at https://groups.google.com/group/votca. >> >>> >>> > For more options, visit https://groups.google.com/d/optout. >> >>> >>> >> >>> >>> >> >>> >>> >> >>> >>> -- >> >>> >>> Christoph Junghans >> >>> >>> Web: http://www.compphys.de >> >>> >>> >> >>> >>> -- >> >>> >>> You received this message because you are subscribed to the Google >> >>> >>> Groups >> >>> >>> "votca" group. >> >>> >>> To unsubscribe from this group and stop receiving emails from it, >> >>> >>> send an >> >>> >>> email to [email protected]. >> >>> >>> To post to this group, send email to [email protected]. >> >>> >>> Visit this group at https://groups.google.com/group/votca. >> >>> >>> For more options, visit https://groups.google.com/d/optout. >> >>> >> >> >>> >> >> >>> >> >> >>> >> >> >>> >> -- >> >>> >> Best regards, >> >>> >> Changwoon Jang, >> >>> >> >> >>> >> Postdoctoral Research Fellow >> >>> >> Department of Chemical & Biological Engineering, Drexel University >> >>> >> 3141 Chestnut Street, Philadelphia, PA 19104 >> >>> >> >> >>> >> Voice: (662) 617-2267 >> >>> >> E-mail: [email protected] >> >>> >> >> >>> >> -- >> >>> >> You received this message because you are subscribed to the Google >> >>> >> Groups >> >>> >> "votca" group. >> >>> >> To unsubscribe from this group and stop receiving emails from it, >> >>> >> send >> >>> >> an >> >>> >> email to [email protected]. >> >>> >> To post to this group, send email to [email protected]. >> >>> >> Visit this group at https://groups.google.com/group/votca. >> >>> >> For more options, visit https://groups.google.com/d/optout. >> >>> > >> >>> > >> >>> > >> >>> > -- >> >>> > Christoph Junghans >> >>> > Web: http://www.compphys.de >> >>> >> >>> >> >>> >> >>> -- >> >>> Christoph Junghans >> >>> Web: http://www.compphys.de >> >>> >> >>> -- >> >>> You received this message because you are subscribed to the Google >> >>> Groups >> >>> "votca" group. >> >>> To unsubscribe from this group and stop receiving emails from it, send >> >>> an >> >>> email to [email protected]. >> >>> To post to this group, send email to [email protected]. >> >>> Visit this group at https://groups.google.com/group/votca. >> >>> For more options, visit https://groups.google.com/d/optout. >> >> >> >> >> >> >> >> >> >> -- >> >> Best regards, >> >> Changwoon Jang, >> >> >> >> Postdoctoral Research Fellow >> >> Department of Chemical & Biological Engineering, Drexel University >> >> 3141 Chestnut Street, Philadelphia, PA 19104 >> >> >> >> Voice: (662) 617-2267 >> >> E-mail: [email protected] >> > >> > >> > >> > >> > -- >> > Best regards, >> > Changwoon Jang, >> > >> > Postdoctoral Research Fellow >> > Department of Chemical & Biological Engineering, Drexel University >> > 3141 Chestnut Street, Philadelphia, PA 19104 >> > >> > Voice: (662) 617-2267 >> > E-mail: [email protected] >> > >> > -- >> > You received this message because you are subscribed to the Google >> > Groups >> > "votca" group. >> > To unsubscribe from this group and stop receiving emails from it, send >> > an >> > email to [email protected]. >> > To post to this group, send email to [email protected]. >> > Visit this group at https://groups.google.com/group/votca. >> > For more options, visit https://groups.google.com/d/optout. >> >> >> >> -- >> Christoph Junghans >> Web: http://www.compphys.de >> >> -- >> You received this message because you are subscribed to the Google Groups >> "votca" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> To post to this group, send email to [email protected]. >> Visit this group at https://groups.google.com/group/votca. >> For more options, visit https://groups.google.com/d/optout. > > > > > -- > Best regards, > Changwoon Jang, > > Postdoctoral Research Fellow > Department of Chemical & Biological Engineering, Drexel University > 3141 Chestnut Street, Philadelphia, PA 19104 > > Voice: (662) 617-2267 > E-mail: [email protected] > > -- > You received this message because you are subscribed to the Google Groups > "votca" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at https://groups.google.com/group/votca. > For more options, visit https://groups.google.com/d/optout. -- Christoph Junghans Web: http://www.compphys.de -- You received this message because you are subscribed to the Google Groups "votca" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/votca. For more options, visit https://groups.google.com/d/optout.
