On Thursday, April 12, 2018 at 4:55:47 PM UTC-4, Christoph Junghans wrote: > > On Thu, Apr 12, 2018 at 2:28 PM, Alexander Alexander > <[email protected] <javascript:>> wrote: > > > > Thank you. > > > > An then these parameter > > <pot_max>10000000</pot_max> > > <table_end>4.0</table_end> > > which usually comes in the very end of settings.xml of IBI in the > <inverse> > > section, are again mx of potential and end of it, which should be larger > > than all our end table in each section. > Like I said, these are parameter for the gromacs table (the xvg file), > not the VOTCA table (the pot.new file). > You could set them differently for each table, but usually the values > above should work for most Gromacs runs. > > > > > On Thursday, April 12, 2018 at 4:09:27 PM UTC-4, Christoph Junghans > wrote: > >> > >> On Thu, Apr 12, 2018 at 2:03 PM, Alexander Alexander > >> <[email protected]> wrote: > >> > > >> > Hi Christoph, > >> > > >> > I guess I found out what do you mean, I hope the attached bonded > tables > >> > for > >> > 3-bead system would be fine. > >> Yeah these one look better, if you zoom around the minimum you should > >> see some structure for the Boltzmann inversion. > > > > > > Yes, when I zoom in them, I see a wide U shape for each of bonded > > interactions as in attachment. > > Now, to use them in IBI (only nonbonded), should I use the zoomed > version of > > them or these are fine for IBI over nonbonded? > If you are doing only non-bonded IBI, just provide VOTCA with the full > table_a1.xvg, table_b1.xvg etc. and add them to <filelist> as well. > Also don't forget to remove the <bonded> sections of the interactions > you aren't iterating from your settings.xml file. > Probably I can even keep them in the setting but not updating them, I mean "<do_potential>0 0 0</do_potential>", but improving them via below command just by smoothing and scaling them as below. <post_update>smooth</post_update> <post_update_options> <scale>0.001</scale> </post_update_options> <post_add>convergence</post_add> How about if I want to do the full IBI (both bonded and nonbonded), using those table_a,b.xvg?
Regards, Alex > > Christoph > > > > Thank you. > > Regards, > > Alex > >> > >> > >> Christoph > >> > > >> > Best regards, > >> > Alex > >> > > >> > > >> > On Thursday, April 12, 2018 at 1:26:37 PM UTC-4, Alexander Alexander > >> > wrote: > >> >> > >> >> > >> >> Thanks. > >> >> > >> >> On Thursday, April 12, 2018 at 12:43:54 PM UTC-4, Christoph Junghans > >> >> wrote: > >> >>> > >> >>> On Thu, Apr 12, 2018 at 9:57 AM, Alexander Alexander > >> >>> <[email protected]> wrote: > >> >>> > > >> >>> > > >> >>> > On Thursday, April 12, 2018 at 8:14:42 AM UTC-4, Christoph > Junghans > >> >>> > wrote: > >> >>> >> > >> >>> >> On Thu, Apr 12, 2018 at 3:39 AM, Alexander Alexander > >> >>> >> <[email protected]> wrote: > >> >>> >> > > >> >>> >> > Hi, > >> >>> >> > > >> >>> >> > I was wondering if it is normal to have table_b1,2,3,4.xvg > like > >> >>> >> > the > >> >>> >> > attached > >> >>> >> > figures? They are the result if BI (similar script as hexane > >> >>> >> > example) > >> >>> >> > over a > >> >>> >> > 500 ns nvt simulation of a single molecule. What could be the > >> >>> >> > meaning of > >> >>> >> > the > >> >>> >> > straight line passing from point (0,0) in table_b1.xvg for > >> >>> >> > example? > >> >>> >> No they don't look normal, bonded interaction should always look > >> >>> >> like > >> >>> >> an modified U with both sides, left and right, going to > infinity. > >> >>> > > >> >>> > > >> >>> > I do not know what exactly I am doing wrong; I guess the 500 ns > NVT > >> >>> > simulation is fine to get statistic for a single molecule, and > then > >> >>> > I > >> >>> > am > >> >>> > following these command in the hexane example for to obtain > >> >>> > bonded.xvg > >> >>> > potential using BI: > >> >>> > > >> >>> > #!/bin/bash -e > >> >>> > > >> >>> > #calculate bonded potentials with csg_boltzmann > >> >>> > > >> >>> > csg_boltzmann --top prdnvt.tpr --trj prdnvt.trr --cg mapp.xml > >> >>> > --begin > >> >>> > 2000 < > >> >>> > boltzmann_cmds > >> >>> > > >> >>> > #smooth bonded potentials > >> >>> > csg_call --sloppy-tables table smooth bond-LG-H1.pot.ib > >> >>> > input_bond-LG-H1.pot > >> >>> > csg_call --sloppy-tables table smooth bond-H1-H2.pot.ib > >> >>> > input_bond-H1-H2.pot > >> >>> > csg_call --sloppy-tables table smooth bond-H2-L.pot.ib > >> >>> > input_bond-H2-L.pot > >> >>> > csg_call --sloppy-tables table smooth bond-L-RG.pot.ib > >> >>> > input_bond-L-RG.pot > >> >>> > csg_call --sloppy-tables table smooth angle-LG-H1-H2.pot.ib > >> >>> > input_angle-LG-H1-H2.pot > >> >>> > csg_call --sloppy-tables table smooth angle-H1-H2-L.pot.ib > >> >>> > input_angle-H1-H2-L.pot > >> >>> > csg_call --sloppy-tables table smooth angle-H2-L-RG.pot.ib > >> >>> > input_angle-H2-L-RG.pot > >> >>> > csg_call --sloppy-tables table smooth dihedral-LG-H1-H2-L.pot.ib > >> >>> > input_dihedral-LG-H1-H2-L.pot > >> >>> > csg_call --sloppy-tables table smooth dihedral-H1-H2-L-RG.pot.ib > >> >>> > input_dihedral-H1-H2-L-RG.pot > >> >>> > > >> >>> > #convert bonded potentials to GROMACS tables > >> >>> > if [ -d table ]; then > >> >>> > rm -r table > >> >>> > fi > >> >>> > mkdir table > >> >>> > > >> >>> > csg_call --ia-type bond --ia-name bond-LG-H1 --options > >> >>> > bond-LG-H1.xml > >> >>> > convert_potential gromacs --clean input_bond-LG-H1.pot > >> >>> > table/table_b1.xvg > >> >>> > csg_call --ia-type bond --ia-name bond-H1-H2 --options > >> >>> > bond-H1-H2.xml > >> >>> > convert_potential gromacs --clean input_bond-H1-H2.pot > >> >>> > table/table_b2.xvg > >> >>> > csg_call --ia-type bond --ia-name bond-H2-L --options > bond-H2-L.xml > >> >>> > convert_potential gromacs --clean input_bond-H2-L.pot > >> >>> > table/table_b3.xvg > >> >>> > csg_call --ia-type bond --ia-name bond-L-RG --options > bond-L-RG.xml > >> >>> > convert_potential gromacs --clean input_bond-L-RG.pot > >> >>> > table/table_b4.xvg > >> >>> > csg_call --ia-type angle --ia-name angle-LG-H1-H2 --options > >> >>> > angle-LG-H1-H2.xml convert_potential gromacs --clean > >> >>> > input_angle-LG-H1-H2.pot table/table_a1.xvg > >> >>> > csg_call --ia-type angle --ia-name angle-H1-H2-L --options > >> >>> > angle-H1-H2-L.xml > >> >>> > convert_potential gromacs --clean input_angle-H1-H2-L.pot > >> >>> > table/table_a2.xvg > >> >>> > csg_call --ia-type angle --ia-name angle-H2-L-RG --options > >> >>> > angle-H2-L-RG.xml > >> >>> > convert_potential gromacs --clean input_angle-H2-L-RG.pot > >> >>> > table/table_a3.xvg > >> >>> > csg_call --ia-type angle --ia-name dihedral-LG-H1-H2-L --options > >> >>> > dihedral-LG-H1-H2-L.xml convert_potential gromacs --clean > >> >>> > input_dihedral-LG-H1-H2-L.pot table/table_d1.xvg > >> >>> > csg_call --ia-type angle --ia-name dihedral-H1-H2-L-RG --options > >> >>> > dihedral-H1-H2-L-RG.xml convert_potential gromacs --clean > >> >>> > input_dihedral-H1-H2-L-RG.pot table/table_d2.xvg > >> >>> > > >> >>> > Even I have similar problem for a molecule with three beads. > >> >>> > Attachment is a log.txt file out of running the above script. Any > >> >>> > help > >> >>> > would > >> >>> > be highly appreciated. > >> >>> As discussed on the mailing list in great detail, the art is to > pick > >> >>> the min and max of the potentials, so that the extrapolation is > >> >>> useful. > >> >> > >> >> You mean I have to edit (clip) the pot.in I get from csg_boltzmann > (tab > >> >> ....) (the first command in the above script), as bond-case.pot.in, > >> >> right? > >> >> Actually, that is the point I am getting confused. > >> >> There are several min and max set in input files: > >> >> One min and max set is in csg_boltzmann such as: > >> >> tab set min XXX > >> >> tab set max YYY > >> >> > >> >> One is in one set of min max in the <bonded> section of bond-A-B.xml > >> >> feed > >> >> to the csg_call. like: > >> >> <bonded> > >> >> <min>xxx</min> > >> >> <max>yyy</max> > >> >> <step>0.001</step> > >> >> </bonded> > >> >> Again there is another set of min max in the <inverse> section of > the > >> >> same > >> >> file as below: > >> >> <inverse> > >> >> <gromacs> > >> >> <pot_max>1e8</pot_max> > >> >> <table_end>0.4</table_end> > >> >> <table_bins>0.001</table_bins> > >> >> </gromacs> > >> >> </inverse> > >> >> > >> >>> You can basically not expect VOTCA will get this right > automatically. > >> >>> And nothing prevents you from doing this by hand, right? You are > >> >>> right. > >> >>> And then you can provide your hand-tuned initial guess as a pot.in > >> >>> file > >> >>> to IBI. > >> >> > >> >> Probably you mean rest of BI (like csg_call and converting to > >> >> table_*.xvg > >> >> format), because I still getting the bonded potentials using BI. And > >> >> the > >> >> final table_*.xvg are supposed be used as input for IBI. > >> >> > >> >> Best regards, > >> >> Alex > >> >>> > >> >>> > >> >>> Christoph > >> >>> > >> >>> > > >> >>> > BTW, how the form of the angle-bonded potential should look like? > >> >>> > The > >> >>> > table_a1.xvg for hexane does not go to +infinity from the right > leg. > >> >>> > Thank you. > >> >>> > Regards, > >> >>> > Alex > >> >>> > > >> >>> >> > >> >>> >> > Also, is the format of the table_bonded.xvg is "x y [error > bar]"? > >> >>> >> x,y, force - see the gromacs manual for details. > >> >>> >> > >> >>> >> Christoph > >> >>> >> > >> >>> >> > Thank you. > >> >>> >> > Regards, > >> >>> >> > Alex > >> >>> >> > > >> >>> >> > On Wednesday, April 11, 2018 at 3:44:30 PM UTC-4, Christoph > >> >>> >> > Junghans > >> >>> >> > wrote: > >> >>> >> >> > >> >>> >> >> On Wed, Apr 11, 2018 at 12:49 PM, Alexander Alexander > >> >>> >> >> <[email protected]> wrote: > >> >>> >> >> > > >> >>> >> >> > Hi Christoph, > >> >>> >> >> > > >> >>> >> >> > On Wednesday, April 11, 2018 at 8:22:24 AM UTC-4, Christoph > >> >>> >> >> > Junghans > >> >>> >> >> > wrote: > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> On Wed, Apr 11, 2018, 01:51 Alexander Alexander > >> >>> >> >> >> <[email protected]> > >> >>> >> >> >> wrote: > >> >>> >> >> >>> > >> >>> >> >> >>> > >> >>> >> >> >>> Hi, > >> >>> >> >> >>> You mean BI over a single molecule in vacuum? > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> Yes, or any potential, which has right equilibrium > >> >>> >> >> >> length/value. > >> >>> >> >> > > >> >>> >> >> > I decided to the BI for bonded interaction, the procedure > and > >> >>> >> >> > results > >> >>> >> >> > for > >> >>> >> >> > one of the molecule is shared in the below link, I am not > sure > >> >>> >> >> > if > >> >>> >> >> > everything > >> >>> >> >> > is fine, I would be so appreciated if you could kindly have > a > >> >>> >> >> > look > >> >>> >> >> > and > >> >>> >> >> > give > >> >>> >> >> > any suggestion, please. > >> >>> >> >> > > >> >>> >> >> > > >> >>> >> >> > > >> >>> >> >> > > https://drive.google.com/open?id=1hvIH9mk1LfYo5LOO1VjBAkBZx7L0q3P0 > >> >>> >> >> > > >> >>> >> >> > Also, do the settings used in the hexane or spce water > example > >> >>> >> >> > in > >> >>> >> >> > the > >> >>> >> >> > tutorial are the real settings and parameters used in your > >> >>> >> >> > publication? > >> >>> >> >> > or > >> >>> >> >> > these are just examples. > >> >>> >> >> Most of them are, but some got adapted a bit to be more > >> >>> >> >> instructive > >> >>> >> >> as > >> >>> >> >> tutorials. > >> >>> >> >> > >> >>> >> >> Christoph > >> >>> >> >> > > >> >>> >> >> > Thanks. > >> >>> >> >> > Regards, > >> >>> >> >> > Alex > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >>> > >> >>> >> >> >>> Instead of sd integrater can md integrator (but still in > NVT > >> >>> >> >> >>> ensemble) > >> >>> >> >> >>> be > >> >>> >> >> >>> used in IBI ( bonded and nonbonded or either of them) and > >> >>> >> >> >>> still > >> >>> >> >> >>> have > >> >>> >> >> >>> good > >> >>> >> >> >>> results? > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> Sure as long as you are still sampling NVT you can use any > >> >>> >> >> >> integrator. > >> >>> >> >> >> However, usually sd is the more stable integrator, so I > don't > >> >>> >> >> >> think > >> >>> >> >> >> that > >> >>> >> >> >> change will help with your problem. > >> >>> >> >> >> > >> >>> >> >> >> Christoph > >> >>> >> >> >> > >> >>> >> >> >>> Thanks, > >> >>> >> >> >>> 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. > >> >>> >> >> > > >> >>> >> >> > -- > >> >>> >> >> > 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. > >> > >> > >> > >> -- > >> 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] <javascript:>. > > To post to this group, send email to [email protected] > <javascript:>. > > 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.
