понедельник, 16 декабря 2013 г., 20:04:38 UTC+4 пользователь Christoph Junghans написал: > > 2013/12/16 Sergey Larin <[email protected] <javascript:>> > > > > Dear Sikandar, > > > > Thank you for your help. I've done everything you describe already. I've > prepared mapping and topology files and start test IBI process. I'll try to > describe problem that exists, from my point of view in more details. > > Let's use notation proposed by you and see at A-B distribution. Monomer > unit of my polymer consists of 11 beads that have 4 different types: > -C1-D1-A1-D1-C1-B1-A2-B2-B3-A3-B4- > > This topology is described in VOTCA mapping file that I use to calculate > distributions with csg_stat. Of course, it excludes interactions for beads > bonded directly. for example, A2-B2 and B3-A3 pairs are not included in > distribution. This is right. But it should also exclude pairs in which > beads are connected by two consecutive bonds, for example, A2-B3 and B2-A3, > but csg_stat does not do this exclusion. That is why I have the peak on the > A-B pair distribution function. And this is a problem. > > As far as I know, Gromacs does not calcuate non-bonded interactions for > beads connected by one, two and three consecutive bonds, thus only > interactions between pairs B1-A3 and A2-B4 are calculated during > simulation. Thus, only these two pairs should be taken into account when > calculation distribution, and csg_stat does not do this. > > > > So, my question is how to exclude 1-3 and 1-4 pairs from distributions > calculated by csg_stat? Is it possible at all? Unfortunately I've not found > any solutions neither in manual, nor in tutorial samples. > VOTCA has no way to define the number of consecutive bonds to exclude, > but you can create a couple of extra bonds and angles to achieve the > same number of exclusions. > > O'K. Thanks. Could you say which atoms are excluded by csg_stat? Connencted by bonds and included in angles and dihedrals? If it so, I'll define additional dihedral angles in my simulation and it should solve the problem.
Sergey Cheers, > > Christoph > > > > Sincerely, > > Sergey > > > > > > суббота, 14 декабря 2013 г., 0:41:51 UTC+4 пользователь sikandar > написал: > >> > >> Hi Sergey > >> > >> From your description of the problem, I understand that you want to use > csg_stat to compute > >> only the non-bonded distributions between specified atoms. Right? > >> > >> If so.. csg_stat can be used to compute both bonded and non-bonded > distributions separately. > >> To achieve that one approach would be to specify your molecule topology > that each atom has > >> unique and different name though there may be multiple atoms of same > type. In your monomer case, > >> there are four different types of atoms A, B, C, D, and each monomer > consists of 2 As, 3 Bs, 2Cs, and 2 Ds. > >> Then, in topology, you can specify atom types as A, B, C, D and atom > names are A1, A2, B1, B2, B3, C1, C2, > >> and D1, and D2. For clarification, you can check out the propane > example in votca-tutorials. > >> > >> Now with above mentioned topolopy, you can compute non-bonded > distributions between A1-A2, C1-C2, and so on. > >> > >> Next, if you want to perform coarse-graining and need a reference pair > distributions then first you need to define > >> a mapping to specify how to determine CG beads configurations from a > given reference all-atom configuration. Then > >> you can provide this mapping file to csg_stat to determine target > distributions for coarse-graining. > >> Again you can check out propane example and votca manual to get more > information about how to achieve this. > >> > >> I hope this helps. > >> > >> Cheers > >> Sikandar > >> > >> > >> On Fri, Dec 13, 2013 at 3:36 AM, Sergey Larin <[email protected]> > wrote: > >>> > >>> Dear votca users, > >>> > >>> I've faced one more problem (not so big, I suppose) in coarsegrainig > using votca. I've found that csg_stat calculate pair distribution functions > taking into account atoms that are bonded. As a result the distributions > have peaks that correspond to bonded interaction (see the picture, in the > molecule I'm coarsegraining the monomer unit is -C-D-A-D-C-A-B-B-A-B-). It > is clear for me that it is pretty useless to perform IBI for non-bonded > interactions using these distribution as target ones. Thus, my question is > how to say csg_stat to exclude bonded atoms from distribution caclulations? > Unfortunately, I've found only the way to create exclusion list for > non-bonded interactions to perform reference simulations for bonded > interactions distributions. > >>> > >>> Thanks in advance. > >>> > >>> Sincerely, > >>> Sergey > >>> > >>> > >>> -- > >>> 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 http://groups.google.com/group/votca. > >>> For more options, visit https://groups.google.com/groups/opt_out. > >> > >> > > -- > > 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] <javascript:>. > > To post to this group, send email to [email protected]<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 [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/votca. For more options, visit https://groups.google.com/groups/opt_out.
