Hi, Off topic for this discussion, but what would people think if we changed things around so that grompp would do that constraining, before writing the tpr file?
This will mean that the configuration in the tpr does not always match the configuration that was given to grompp, but instead matches the rest of the simulation. Thus such an energy jump would not occur. And it may mean that broken structures are found at grompp time rather than later at step 0 when doing mdrun on the cluster. Mark On Fri., 29 Mar. 2019, 02:17 Justin Lemkul, <[email protected]> wrote: > > > On 3/28/19 6:02 PM, Kruse, Luke E.(MU-Student) wrote: > > In the production run I have constraints on h-bonds and I do not have > any constraints in the minimization. > > This would account for a sudden change in energy. At step 0, your H > positions will immediately change. > > -Justin > > > ________________________________ > > From: [email protected] < > [email protected]> on behalf of David van > der Spoel <[email protected]> > > Sent: Thursday, March 28, 2019 4:50:38 PM > > To: [email protected] > > Subject: Re: [gmx-users] Energy Conservation at the Beginning of a > Production Run > > > > Den 2019-03-28 kl. 21:19, skrev Kruse, Luke E.(MU-Student): > >> In the mdout.mdp file corresponding to this run I have the line > >> > >> ; Do not constrain the start configuration > >> continuation = no > >> > >> > >> Do I need to change this to yes? > > Maybe, try a 100 step run where you save energies at each time step. Did > > you change anything else between minimization and production, e.g. > > constraint settings? > >> > >> The plot of energy vs time looks like an under damped response to a set > point change in control theory - slight overshooting after the first few > steps, then settling down to an average value nicely. This is what is > leading me to believe that there is a large, non conservative force near > the beginning that becomes smaller, and conservative. > >> > >> Time (ps) > >> > >> 0.000000 -1851396.981075 > >> 0.100000 -1851388.994856 > >> 0.200000 -1851390.001591 > >> 0.300000 -1851389.436699 > >> 0.400000 -1851389.760231 > >> 0.500000 -1851389.726310 > >> 0.600000 -1851389.808428 > >> 0.700000 -1851389.907829 > >> 0.800000 -1851389.840644 > >> 0.900000 -1851389.967373 > >> 1.000000 -1851390.061502 > >> 1.100000 -1851390.057500 > >> 1.200000 -1851390.086172 > >> 1.300000 -1851390.042747 > >> 1.400000 -1851390.060210 > >> 1.500000 -1851389.998599 > >> 1.600000 -1851390.120319 > >> 1.700000 -1851390.141846 > >> 1.800000 -1851390.131003 > >> 1.900000 -1851390.170220 > >> 2.000000 -1851390.141617 > >> 2.100000 -1851390.135779 > >> 2.200000 -1851390.255565 > >> 2.300000 -1851390.153947 > >> 2.400000 -1851390.238681 > >> 2.500000 -1851390.144512 > >> 2.600000 -1851390.255200 > >> 2.633000 -1851390.178012 > >> > >> > >> ________________________________ > >> From: [email protected] < > [email protected]> on behalf of David van > der Spoel <[email protected]> > >> Sent: Thursday, March 28, 2019 3:10:05 PM > >> To: [email protected] > >> Subject: Re: [gmx-users] Energy Conservation at the Beginning of a > Production Run > >> > >> Den 2019-03-28 kl. 20:53, skrev Kruse, Luke E.(MU-Student): > >>> Hello gromacs users, > >>> > >>> > >>> I am trying to simulate a peptide amphiphile with the CHARMM27 force > field. To do this I have had to specify an additional bond type and bond, > angle, dihedral, etc. parameters in the .itp files of the force field. > Then, to check if I had done this correctly, I minimized the system, and > ran a production run to generate an NVE ensemble, so that I could make sure > the system was conserving energy appropriately. After looking at the energy > vs time plot (produces with the gmx energy command), however, the system > jumps from an initial energy, up ~20 kJ per mole and then conserves energy > for the most part (slight drifting but seems tolerable relative to the > initial discontinuity). Is this discontinuity a normal happenstance or a > result of bad minimization? > >>> > >> Is that at step 0 in the simulation? > >> > >> Please check options like > >> ; Do not constrain the start configuration > >> continuation = no > >> > >> > >>> Thank you! > >>> > >>> Luke > >>> > >> > >> -- > >> David van der Spoel, Ph.D., Professor of Biology > >> Head of Department, Cell & Molecular Biology, Uppsala University. > >> Box 596, SE-75124 Uppsala, Sweden. Phone: +46184714205. > >> http://www.icm.uu.se > >> -- > >> Gromacs Users mailing list > >> > >> * Please search the archive at > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before > posting! > >> > >> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists > >> > >> * For (un)subscribe requests visit > >> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or > send a mail to [email protected]. > >> > > > > -- > > David van der Spoel, Ph.D., Professor of Biology > > Head of Department, Cell & Molecular Biology, Uppsala University. > > Box 596, SE-75124 Uppsala, Sweden. Phone: +46184714205. > > http://www.icm.uu.se > > -- > > Gromacs Users mailing list > > > > * Please search the archive at > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before > posting! > > > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists > > > > * For (un)subscribe requests visit > > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or > send a mail to [email protected]. > > -- > ================================================== > > Justin A. Lemkul, Ph.D. > Assistant Professor > Office: 301 Fralin Hall > Lab: 303 Engel Hall > > Virginia Tech Department of Biochemistry > 340 West Campus Dr. > Blacksburg, VA 24061 > > [email protected] | (540) 231-3129 > http://www.thelemkullab.com > > ================================================== > > -- > Gromacs Users mailing list > > * Please search the archive at > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before > posting! > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists > > * For (un)subscribe requests visit > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or > send a mail to [email protected]. > -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to [email protected].
