W dniu poniedziałek, 1 lipca 2013 16:27:04 UTC+1 użytkownik Christoph Junghans napisał: > > 2013/7/1 <s.neu...@gmail.com <javascript:>>: > > > > > > W dniu poniedziałek, 1 lipca 2013 15:28:35 UTC+1 użytkownik Christoph > > Junghans napisał: > >> > >> 2013/7/1 <s.neu...@gmail.com>: > >> > > >> > > >> > W dniu poniedziałek, 1 lipca 2013 15:04:38 UTC+1 użytkownik Christoph > >> > Junghans napisał: > >> >> > >> >> 2013/7/1 <s.neu...@gmail.com>: > >> >> > > >> >> > > >> >> > W dniu poniedziałek, 1 lipca 2013 02:51:23 UTC+1 użytkownik > Christoph > >> >> > Junghans napisał: > >> >> >> > >> >> >> 2013/6/18 <s.neu...@gmail.com>: > >> >> >> > Dear Votca Users, > >> >> >> > > >> >> >> > I have started using the software some time ago and i > successfully > >> >> >> > managed > >> >> >> > to extract bonded parameters from atomistic into the > >> >> >> > coarse-grained - > >> >> >> > this > >> >> >> > is done. > >> >> >> > i have my peptide made of 90 residues - each residue is one > bead, > >> >> >> > however I > >> >> >> > wish to have nonbonded parameters for 5 types of beads (A, B, > C, > >> >> >> > D, > >> >> >> > E). > >> >> >> > In > >> >> >> > this case I will group my amino acids into given types of > beads. > >> >> >> > > >> >> >> > As I can see the input for IBI method is a RDF e.g. A-A, A-B, > B-B, > >> >> >> > etc > >> >> >> > (25 > >> >> >> > in total). From atomistic I will get 90*90/2 = 4050 RDF between > >> >> >> > residues. In > >> >> >> > order to get an input for IBI shall I take an average of of RDF > >> >> >> > for > >> >> >> > A-A, > >> >> >> > A-B, etc (25 in total) from atomistic (4050 in total)? It will > not > >> >> >> > reproduce > >> >> >> > atomistic in this case of its average. Please, explain and > advice > >> >> >> > me. > >> >> >> I guess you need to average your 4050 rdf to get 25 averaged > rdfs. > >> >> >> > >> >> >> You could do that with csg_stat or g_rdf if your simulation was > >> >> >> done > >> >> >> with gromacs. > >> >> >> > >> >> >> Christoph > >> >> >> > >> >> >> PS: Sorry for the delay, this email slipped under my radar. > >> >> > > >> >> > > >> >> > Thank you. I think I need to calculate rdf between different types > >> >> > specified > >> >> > in Gromacs... however, residues are neighboring so I wish to > exclude > >> >> > those > >> >> > interactions. In gromacs g_rdf does not have this option (-cut > option > >> >> > will > >> >> > just cut the distance which is incorrect). Would g_stat be able to > >> >> > exclude > >> >> > neighbors from my atomistic? > >> >> For g_rdf, you could use a special index file or create a tpr with > >> >> additional exclusion. > >> >> > >> >> In csg_stat you could add bonds (in the mapping xml file) between > the > >> >> beads, which you want to exclude. > >> >> > >> >> > Another issue: I used BI for angles - I tried refining my > potential > >> >> > using > >> >> > VOTCA but it somehow describes it poorly... Does csg_resample use > a > >> >> > function > >> >> > to refine it? What kind of function? > >> >> csg_resample uses akima spline by default, but it can also do cubic > >> >> spline or linear interpolation. > >> >> > >> >> > I would prefer to refine it my own but no clue how I will then > >> >> > calculate > >> >> > the > >> >> > derivative of those values (3rd column in table)a*.xvg). Would you > >> >> > advise? > >> >> use csg_resample with --type linear and --derivative file option. > >> >> > >> >> Christoph > >> > > >> > > >> > Thank you a lot! csg_resample --type linear will use linear function > >> > which > >> > is not desired... My distributions are made of multiple gaussian > >> > distributions so i guess akima (deafult) option? > >> If you are using the same grid for input and output, the y points > won't > >> change. > >> Linear interpolation will just draw lines between the given points. > >> If you have distributions made of multiple Gaussian, why don't you fit > >> multiple Gaussians to it and write the result out as a table again. > > > > > > Thank you. I tried to fit in another software my potential to 6th order > > polynomial but results are not promising... which option of csg_stat > will > > fit it to multiple gaussian? > There is no VOTCA tools doing that, but you could use scipy or gnuplot. > > >> > >> > >> > > >> > Does Votca 1.2.3 supports IBI for bonded or shall I upgarde it to 1.3 > ? > >> Bonded IBI only works in Votca 1.3! > See: > <https://code.google.com/p/votca/wiki/Installing#Development_Version> > > > > > > > I will upgrade it. For bonded do we provide both distributions and > > potentials in settings? > > > > <cg> > > <bonded> > > <name>angle</name> > > <min>1</min> > > <max>3</max> > > <step>0.05</step> > > <inverse> > > <target>angle.dist1.tgt</target> > > <gromacs> > > <table>table_a1.xvg</table> > > </gromacs> > > </inverse> > > </bonded> > > > > So the input are both refined table_a1.xvg as well as angle.dist1.tgt? > Will > > they be updated every step? > angle.dist1.tgt is the target distribution against which csg_inverse > compares the current distribution, it is never updated. > table_a1.xvg is the table to export VOTCAs internal potential to, it > is updated every iteration step. > > Thank you. But for both bonded and nonbonded do I have to have two inputs: distribution and potential? Another issue: are tables*xvg copied every time or shall I add them to the list to be copied? they are not specified in the tutorial as files to be copied:
http://code.google.com/p/votca/source/browse/hexane/ibi_bonded/settings.xml?repo=csg-tutorials Steven > > > > Thanks, > > > > Steven > > > >> > > >> > > >> >> > >> >> > > >> >> > Steven > >> >> > > >> >> > > >> >> >> > >> >> >> > >> >> >> > Steven > >> >> >> > > >> >> >> > -- > >> >> >> > 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 votca+un...@googlegroups.com. > >> >> >> > To post to this group, send email to vo...@googlegroups.com. > >> >> >> > Visit this group at http://groups.google.com/group/votca. > >> >> >> > For more options, visit > https://groups.google.com/groups/opt_out. > >> >> >> > > >> >> >> > > >> >> >> > >> >> >> > >> >> >> > >> >> >> -- > >> >> >> 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 votca+un...@googlegroups.com. > >> >> > To post to this group, send email to vo...@googlegroups.com. > >> >> > Visit this group at http://groups.google.com/group/votca. > >> >> > For more options, visit https://groups.google.com/groups/opt_out. > >> >> > > >> >> > > >> >> > >> >> > >> >> > >> >> -- > >> >> 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 votca+un...@googlegroups.com. > >> > To post to this group, send email to vo...@googlegroups.com. > >> > Visit this group at http://groups.google.com/group/votca. > >> > For more options, visit https://groups.google.com/groups/opt_out. > >> > > >> > > >> > >> > >> > >> -- > >> 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 votca+un...@googlegroups.com <javascript:>. > > To post to this group, send email to vo...@googlegroups.com<javascript:>. > > > Visit this group at http://groups.google.com/group/votca. > > For more options, visit https://groups.google.com/groups/opt_out. > > > > > > > > -- > 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 votca+unsubscr...@googlegroups.com. To post to this group, send email to votca@googlegroups.com. Visit this group at http://groups.google.com/group/votca. For more options, visit https://groups.google.com/groups/opt_out.
