Hi, Yes that seems a bit inconsistent. Please file a redmine and include e.g. a tarball of suitable inputs to reproduce what's going on within grompp.
Mark On Tue, Nov 15, 2016 at 3:01 PM Elton Carvalho <[email protected]> wrote: > Hi, Mark, > > Actually, there seems to be a bug somewhere, because if I defined > three type 9 dihedrals in a file and then I define the same three type > 9 dihedrals later on, there should be three warnings: one for each > multiplicity. there were only two warnings, which means the first one > somehow is silent. > > I just did a quick test and re-enabled the first file. The versions > #included by forcefield.itp and by the .top file are identical. I get > two arnings (for multiplicities 2 and 3 like on my first email). If I > comment out these terms, but keep the first one on the second file, I > get no warnings by grompp. > > This means that grompp reads the three terms from file 1 (hiopefully > adding them up), then reads a bunch of parameters and then reads the > first term from file 2, which, according to the documentation (and the > other warnings) would throw a warning and overwrite the prvious > term(s). No warning is given and the user cannot be sure whether the > term was overriden or added up (depending on where the code branches > between these alternatives). > > Could I file a redmine issue regarding that? > > > On Tue, Nov 15, 2016 at 5:02 AM, Mark Abraham <[email protected]> > wrote: > > Hi, > > > > Ok that makes sense. Otherwise it would have seemed like it had to be a > > code bug. > > > > Maybe we could add a warning for including the same file twice? Does > > anybody do that deliberately E.g with different preprocessor variables > > defined? > > > > Mark > > > > On Tue, 15 Nov 2016 00:30 Elton Carvalho <[email protected]> wrote: > > > >> My apologies, > >> > >> After careful inspection, I noticed that I had #included a copy of > >> this ddt.prm file in charmm36-jun2015.ff/forcefield.itp , so the [ > >> dihedraltypes ] directive is actually loaded twice, hence the > >> warnings. Removing the #include from forcefield.itp and keeping it in > >> the .top file eliminated the warnings and the resulting dynamics seems > >> sane (read: did not explode). > >> > >> Lesson learned: don't change the standard forcefield definition files > >> when doing preliminary testing. Especially in the end of a 12h layover > >> between flights. > >> > >> Thank you for your attention, > >> -- > >> Elton > >> > >> On Mon, Nov 14, 2016 at 8:04 PM, Elton Carvalho <[email protected]> > wrote: > >> > Hello, fellow Gromacs Users, > >> > > >> > I'm trying to model P3DDT with CGenFF according to the instructions at > >> > paramchem.org, running the cgenff_charmm2gmx.py script as provided in > >> > their website. All this is being done on gromacs 2016. > >> > > >> > The script converts the .str file generated by paramchem.org to > >> > gromacs syntax and units. As a first test, I ran a signle > >> > dodecylthiophene molecule through the process and obtained apparently > >> > sane .top and .itp files. The .itp file #includes a .prm file, that > >> > contains the parameters fitted by the paramchem software. > >> > > >> > When running gmx grompp (version 2016), I get these two warnings: > >> > > >> > === GROMPP WARNING BEGIN === > >> > > >> > WARNING 1 [file ddt.prm, line 17]: > >> > Overriding Proper Dih. parameters. > >> > Use dihedraltype 9 to allow several multiplicity terms. Only > >> consecutive > >> > lines are combined. Non-consective lines overwrite each other. > >> > > >> > old: 0 0.8368 1 0 0.8368 1 > >> > new: CG2R51 CG2R51 CG321 CG321 9 0.000000 > >> 1.129680 2 > >> > > >> > > >> > WARNING 2 [file ddt.prm, line 18]: > >> > Overriding Proper Dih. parameters. > >> > Use dihedraltype 9 to allow several multiplicity terms. Only > >> consecutive > >> > lines are combined. Non-consective lines overwrite each other. > >> > > >> > old: 0 1.12968 2 0 1.12968 > 2 > >> > new: CG2R51 CG2R51 CG321 CG321 9 0.000000 > >> 0.000000 3 > >> > > >> > === GROMPP WARNING END === > >> > > >> > But the dihedral mentioned is using dihedraltype 9, in lines 16, 17 > >> > and 18 of the reffered file, as follows: > >> > > >> > === ddt.prm RELEVANT SECTION BEGIN (lines 10-21)=== > >> > > >> > [ dihedraltypes ] > >> > ; i j k l func phi0 kphi > mult > >> > CG2R51 CG2R51 CG2R51 CG321 9 180.000000 16.736000 > 2 > >> > CG321 CG2R51 CG2R51 SG2R50 9 180.000000 29.288000 > 2 > >> > CG321 CG2R51 CG2R51 HGR51 9 180.000000 4.184000 > 2 > >> > CG2R51 CG2R51 CG321 CG321 9 0.000000 0.836800 > 1 > >> > CG2R51 CG2R51 CG321 CG321 9 0.000000 1.129680 > 2 > >> > CG2R51 CG2R51 CG321 CG321 9 0.000000 0.000000 > 3 > >> > CG2R51 CG321 CG321 CG321 9 0.000000 0.167360 > 3 > >> > CG2R51 CG321 CG321 HGA2 9 0.000000 0.836800 > 3 > >> > > >> > === ddt.prm RELEVANT SECTION END === > >> > > >> > As you can see, the dihedraltype is 9 and the different multiplicities > >> > of the "CG2R51 CG2R51 CG321 CG321" dihedral are on adjacent > >> > lines (16, 17 and 18 of the original file. Note that the first warning > >> > is about overriding the parameters on line 16 with those on line 17, > >> > so this dihedral is not defined somewhere else). No wildcard dihedral > >> > parameters are present in the loaded .itp files. > >> > > >> > Why does grompp warn me that it's "Overriding Proper Dih. parameters", > >> > even though the parameters are entered (to the best of my knowledge) > >> > as instructed both in the warning and in Section 5.3.3 of the > >> > Reference Manual? How can I check whether it really did override the > >> > parameter or added the different terms? > >> > > >> > Thanks in advance, > >> > -- > >> > Elton Carvalho > >> > >> > >> > >> -- > >> Elton Carvalho > >> -- > >> 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]. > > > > -- > Elton Carvalho > -- > 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].
