Daniel, In regards to extending relative entropy implementation in VOTCA, it is possible but not trivial. Hybrid approach of BI for bonded and Relative entropy for non-bonded can work. So as a starting point, you can use BI for bonded interaction. For extending relative entropy implementation to bonded interactions, see the current implementation in csg source code at src/tools/csg_reupdate.cc . Mainly you would need to implement function void CsgREupdate::AAavgBonded(PotentialInfo* potinfo) . There other things also you need to take care regarding reading, writing bond interaction parameters, and generating bond interaction tables for CG MD simulations.
Thanks, Sikandar On Thu, Feb 1, 2018 at 10:12 AM, Christoph Junghans <[email protected]> wrote: > On Wed, Jan 31, 2018 at 11:14 AM, Adiran <[email protected]> wrote: > > > > > > El miércoles, 31 de enero de 2018, 15:30:03 (UTC), Christoph Junghans > > escribió: > >> > >> On Wed, Jan 31, 2018 at 3:50 AM, <[email protected]> wrote: > >> > El miércoles, 31 de enero de 2018, 5:26:35 (UTC), sikandar escribió: > >> >> > >> >> Hi Daniel, > >> >> > >> >> It is possible to use relative entropy implementation in VOTCA to > >> >> coarse-grain protein. At this point, relative entropy implementation > in > >> >> VOTCA supports only non-bonded interaction coarse-graining. For > bonded > >> >> interaction you will need to use Boltzmann inversion or other > suitable > >> >> method. VOTCA does support multithreaded parallel processing of MD > >> >> trajectories and parallel MD simulations using GROMACS. > >> > > >> > > >> > Is there a plan to include them soon? > >> Not really, but patches are always welcome! > > > > > > Great. Would it be very difficult? > That is more a question for Sikandar! > > > > >> > >> > >> > > >> > So, if I understood correctly, one would create a suitable CG force > >> > field > >> > and then would minimize the non-bonded parameters with RE and the > bonded > >> > ones with Boltzmann inversion in a semi-decoupled way? > > > > > > > >> > >> Yes, do BI first and then use the bonded table in the CG run with RE > >> as fixed interactions. > > > > > > > > Right, so if the AA simulation contains just one protein plus solvent > then > > the parameters would be only optimized for intra-molecular interactions. > > Would these also be transferable to a box with many proteins to simulate, > > for example, aggregation? How does one deal with inter-molecular > > interactions in the VOTCA scheme? > Transferability is very system-dependent, so you will have check that > for your specific system. > > Also sec Section 2.2.1 of the manual <http://doc.votca.org/manual.pdf> > > > > I guess simulating a dimer AA would improve the results but that wouldn't > > still be representative of the real situation. > > > > Also, with respect to the force field: > > > > As I understand it any form of potential (including coulomb) is > acceptable > > as long as you express it as an interpolation with splines. > You can include the coulomb in the tabulated interactions so leave > them separate if you know the CG charges. > > > > > > > > >> > >> > >> > > >> > Also, I am having trouble running the RE tutorials, for example in the > >> > methanol and water one: firstly it doesn't run without me manually > >> > changing > >> > the .mdp file to specify the cutoff scheme to be group and then it is > >> > terminated with: > VOTCA will add the cutoff scheme automatically to the mdp file at > iteration time if it isn't already in the mdp file. > > Did the tutorials without modification work? > > Christoph > >> > > >> > begin to calculate distribution functions > >> > # of bonded interactions: 0 > >> > # of non-bonded interactions: 3 an error occurred: > >> > input format not supported: topol.tpr > >> This error means you have compiled VOTCA without gromacs. Just re-run > >> cmake with -DWITH_GMX=ON and rebuild with "make" again. > > > > > > I did that > > > > I have done that. Now I get a different error (when running the spce RE > > tests) > > > > An error occurred: > > Hessian NOT a positive definite! > > This can be a result of poor initial guess or ill-suited CG potential > > settings or poor CG sampling. > > > > I compiled using: > > > > ./build.sh --prefix ${prefix} --minimal -DWITH_GMX=ON -DWITH_FFTW=ON > > -DWITH_GSL=ON -DGSL_INCLUDE_DIR=/usr/local/Cellar/gsl/2.4/include > > -DGSL_LIBRARY=/usr/local/Cellar/gsl/2.4/lib/libgsl.dylib > > -DCBLAS_LIBRARY=/usr/local/Cellar/gsl/2.4/lib/libgslcblas.dylib > > -DFFTW3_INCLUDE_DIR=/usr/local/Cellar/fftw/3.3.7_1/include > > -DFFTW3_LIBRARY=/usr/local/Cellar/fftw/3.3.7_1/lib/libfftw3.dylib > gromacs > > tools csg > > > > But then that gets translated into: > > > > cmake -DCMAKE_INSTALL_PREFIX='/Users/agaraizar/votca/' > --no-warn-unused-cli > > -DWITH_FFTW=OFF -DWITH_GSL=OFF -DCMAKE_DISABLE_FIND_PACKAGE_GSL=ON > > -DWITH_MKL=OFF -DCMAKE_DISABLE_FIND_PACKAGE_MKL=ON -DBUILD_MANPAGES=OFF > > -DWITH_GMX=OFF -DWITH_SQLITE3=OFF -DWITH_H5MD=OFF > > -DCMAKE_DISABLE_FIND_PACKAGE_HDF5=ON -DWITH_GMX=ON -DWITH_FFTW=ON > > -DWITH_GSL=ON -DGSL_INCLUDE_DIR=/usr/local/Cellar/gsl/2.4/include > > -DGSL_LIBRARY=/usr/local/Cellar/gsl/2.4/lib/libgsl.dylib > > -DCBLAS_LIBRARY=/usr/local/Cellar/gsl/2.4/lib/libgslcblas.dylib > > -DFFTW3_INCLUDE_DIR=/usr/local/Cellar/fftw/3.3.7_1/include > > -DFFTW3_LIBRARY=/usr/local/Cellar/fftw/3.3.7_1/lib/libfftw3.dylib > > -DCMAKE_INSTALL_RPATH_USE_LINK_PATH=ON /Users/agaraizar/votca/src/csg > > > > Thank you > > > >> > >> Christoph > >> > > >> > The GROMACS version I am using is the one provided by VOTCA in the > build > >> > script > >> > > >> > GROMACS: gmx, version 2016.5-dev-20180109-aa8b47218-unknown > >> > Executable: > >> > /Users/agaraizar/votca/bin/gmx > >> > > >> > In the mailing list someone had a similar issue but in her case she > >> > hadn't > >> > installed GROMACS > >> > > >> > Regards, > >> > > >> > Adiran > >> > > >> >> > >> >> For details about the relative entropy implementation and parallel > >> >> performance of VOTCA you can refer to the PLOS One article > >> >> https://doi.org/10.1371/journal.pone.0131754 > >> > > >> > > >> > Thank you > >> > > >> >> > >> >> > >> >> > >> >> --- > >> >> Sikandar > >> >> > >> >> > >> >> > >> >> On Tue, Jan 30, 2018 at 10:26 AM, Christoph Junghans > >> >> <[email protected]> > >> >> wrote: > >> >>> > >> >>> Daniel, > >> >>> > >> >>> I cannot talk to efficiency of the relative entropy method, but in > >> >>> general coarse-graining a protein is hard. You need to have a good > >> >>> intuition how to do the mapping and which interaction to consider. > >> >>> > >> >>> > >> >>> VOTCA has support for running RDF, IMC and relative entropy > >> >>> calculation in parallel. In addition you can speed up the gromacs > runs > >> >>> by using a parallel gromacs as well. > >> >>> > >> > > >> > Thank you for your answer Christoph > >> > > >> >>> > >> >>> Christoph > >> >>> > >> >>> On Tue, Jan 30, 2018 at 5:17 AM, <[email protected]> wrote: > >> >>> > Dear all, > >> >>> > > >> >>> > After reading a lot of the literature available on relative > entropy > >> >>> > minimization as a coarse graining technique I am extremely keen on > >> >>> > using it > >> >>> > for my research. > >> >>> > > >> >>> > I was wondering wether it would be possible/feasible (in a > >> >>> > reasonable > >> >>> > amount > >> >>> > of time) to use VOTCA to coarse grain a protein (~500 atoms) using > >> >>> > RE. > >> >>> > > >> >>> > I am also not sure about what if any parallelization VOTCA has > that > >> >>> > would > >> >>> > reduce the required time (and if it does how to use it). > >> >>> > > >> >>> > Would having a parallel GROMACS help? > >> >>> > > >> >>> > Regards and thanks in advance, > >> >>> > > >> >>> > Daniel > >> >>> > > >> >>> > -- > >> >>> > 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. > >> >> > >> >> > >> > -- > >> > 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. > -- 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.
