On Tue, Apr 10, 2018 at 2:50 PM, Alexander Alexander <[email protected]> wrote: > > Hi Christoph, > > On Tuesday, April 10, 2018 at 4:28:30 PM UTC-4, Christoph Junghans wrote: >> >> On Sun, Apr 8, 2018 at 8:17 PM, Alexander Alexander >> <[email protected]> wrote: >> > >> > Thanks Christoph. >> > >> > I know they are zero, and >> > >> >> <[email protected]> wrote: >> >> > >> >> > Dear all, >> >> > >> >> > The potential_shift.pl crashes in IBI for some of the bond and angles >> >> > in >> >> > step 5, I have already seen a suggestion in mailing list such that >> >> > "one >> >> > should pick the minimum of the interaction (min property) bigger than >> >> > 0, >> >> > so >> >> > that the target rdf is bigger than 0 as well" (although the crash for >> >> > this >> >> > old discussion happened in the step_001). If with minimum of the >> >> > interaction >> >> > you mean is the <min>XXXX</min> in the settings.xml file, then it is >> >> > already >> >> > more that 0 in may case (<min>2.713</min>). Also in general I did not >> >> > understand the solutions proposed over there. So, I have attached >> >> > here >> >> > some >> >> > steps of evolution of the problematic angel during IBI, and I would >> >> > be >> >> > so >> >> > appreciated if one could help overcome the problem. >> >> Looking at your inverse.log: above the error banner "No valid value >> >> found in angle-H-L-RG.dpot.pure_ibi" >> >> Looking at angle-H-L-RG.dpot.pure_ibi, all values are marked "o". >> >> angle-H-L-RG.dist.tgt looks ok, but step_005/angle-H-L-RG.dist.new is >> >> all >> >> zeros. >> >> So why that is I don't know, but you might want to run csg_stat by >> >> hand: >> > >> > Yes, they are all zero. >> > >> >> >> >> csg_stat --nt 32 --options >> >> /home/alex/test/CG/dw/jeffamine-n2/ATB/ibi1/settings.xml --top >> >> topol.tpr --trj tra >> >> j.xtc --begin 50 --first-frame 0 --cg >> >> /home/alex/test/CG/dw/jeffamine-n2/ATB/ibi1/jeffamine2-cg.xml >> > >> > Invoking this show nothing again all zero, while invoking this in >> > step_004 >> > results to a nonzero output. From step 004 to 005 the only thing which >> > change is the traj.xtc, so it might be something with the mdrun over the >> > system. >> My guess is that your system is slowly exploding in step 5 and hence >> there are no value for the angle in range you specified in the xml >> file. > > Thanks for getting back to me, I am getting such an error over an over now > and no success yet just wasting my time. > Indeed, even if do not get the potential_shift error, the system crashes at > some points later in step_008 or so, with an error like "There is no domain > decomposition for" or "bond missing" which are the sign of explosion, even > minimization also do not help. But I am sure that the conf.gro comes from a > very well equlibrated system (around 9 ns of NVT and NPT equilibrations show > a converge density for the system agreeing with experiment), and most > importantly, the MD simulation in each step of CG is in NVT ensemble, so, I > do not know why explosion should happen in an NVT ensemble. > > Would you please let me know the order of potential for a bond like > "bond-RB-RA" in each step? > I mean for this bond a bond-RB-RA.pot.new will be generated in the step_000, > then this bond-A-B.pot.new ---cp--> bond-RB-RA.pot.cur (in step_001) .... . > Or in other word, what is the order of the below potential for this bond, > which is result of which? > bond-RB-RA.dist.new bond-RB-RA.dist.tgt_smooth.ijT3P > bond-RB-RA.pot.extrapol.Rh1pt bond-RB-RA.pot.smooth.sXbTa > bond-RB-RA.dist.tgt bond-RB-RA.dpot.pure_ibi > bond-RB-RA.pot.interpol.0S35s > bond-RB-RA.dist.tgt_extrapolated.Q1VJa bond-RB-RA.pot.cur > bond-RB-RA.pot.shift.V6IRk Just look into inverse.log: $ grep bond-RB-RA inverse.log
> > Also, which version of gromacs you are using? mine is 2018.1 I don't think, this has anything to do with your gromacs version. Try to do IBI on non-bonded interaction only using bonded interactions from simple BI. Christoph > > Best regards, > Alex >> >> >> Christoph >> > >> >> >> >> (from inverse.log) >> >> my guess is that something in wrong in jeffamine2-cg.xml, which should >> >> be a 1:1 mapping on the cg level. >> > >> > If that is the case why this does not happen in the steps before? As >> > they >> > are using the same jeffamine2-cg.xml. >> > Anyway, I have attached both jeffamine2-cg.xml jeffamine2-aa.xml. >> > >> > Best regards, >> > Alex >> >> >> >> Christoph >> >> > Thank you. >> >> > Regards, >> >> > Alex >> >> > >> >> > -- >> >> > 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. >> >> >> >> -- >> 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. -- 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.
