2013/7/8 Zhicheng <[email protected]>:
> Dear Christoph,
> I can't get the parameters by fitting related properties (thermodynamics,
> hydrodynamics) because there are no experimental and theoretical results
> about the coarse-grained beads. Thus, how can I obtain the bonded prameters
> in topology file (e.g. bonds and angles items in propane/ibi/topol.top)
You mean the following part, right?
[ bonds ]
; ai aj funct c0 c1 c2 c3
1 2 8 1 1.0
2 3 8 1 1.0
[ angles ]
; ai aj ak funct c0 c1 c2 c3
1 2 3 8 1 1.0
funct=8 means use a table with number c0=1 and scale it with c1=1.0
(see gromacs manual for details)
The used tables table_a1.xvg (a for angle) and table_b1.xvg (b for
bond) are obtained from a Boltzmann inversion of the atomistic
bond/angle distribution. (see <http://dx.doi.org/10.1021/ct900369w>
for details)
> 在 2013年7月8日星期一UTC+8下午12时45分52秒，Christoph Junghans写道：
>>
>> 2013/7/7 郭志成 <[email protected]>:
>> > Dear Christoph,
>> >
>> > Thanks for your help. It is working now. There is another question about
>> > the
>> > coarse-grained topology file as shown in methanol/ibi/topol.top and
>> > propane/ibi/topol.top. I have been familiar with the atomistic topology.
>> > Therefore, I want to know how to determine these parameters related to
>> > coarse-grained beads (e.g. c6, c12, bonds and angles items).Whether the
>> > specification of these parameters are arbitrary.
>> No, it is not arbitrary, if you are using tables in gromacs the values
>> c6 and c12 are the pre-factors for the table column 4/5 (c6) and 6/7
>> (c12). (also see gromacs manual section 4.2.13)
>> In VOTCA we only use column 6 and 7 for IBI hence we set c12 to 1.0
>> (and c6 to 0, which not 100% necessary as column 4,5 are filled with
>> zeros).
>>
>> In some cases (e.g. spce/tf) it is useful to have c6 and c12 set to
>> something non-trivial.
>>
>> Christoph
>> >
>> >
>> >
>> > Sincerely,
>> >
>> > Zhicheng
>> >
>> >
>> >
>> > 2013/7/8 Christoph Junghans <[email protected]>
>> >>
>> >> 2013/7/6 Zhicheng <[email protected]>:
>> >> > Dear Christoph,
>> >> > I apologize to you for my question. I have known how to get the
>> >> > tabulated
>> >> > potential for bonded and non-bonded interactions.The table.xvg in
>> >> > methanol/ibi you metioned is the file required by gromacs even if
>> >> > there
>> >> > are
>> >> > no other interactions. My question is that how to generate this file.
>> >> table.xvg isn't used anywhere, so you could use a file with zeros
>> >> (like it is done in spce/ibi and the other examples), which I think is
>> >> the safest.
>> >> Where the table.xvg in methanol/ibi cames from, I don't remember, it
>> >> seems to be simple BI potential.
>> >>
>> >> > Whether is it similar to bonded or non-bonded interaction.
>> >> Non-bonded tables have 7 columns while bonded tables have 3 columns.
>> >> Bonded tables usually come from a Boltzmann inversion of the bonded
>> >> distribution.
>> >>
>> >> > In addition, what
>> >> > is the type of interaction represented by table.xvg because that the
>> >> > interaction between the CG beads in methanol/ibi is given as
>> >> > table_CG_CG.xvg.
>> >> Like I said before table.xvg isn't used in this simulation, but
>> >> gromacs uses it for the interaction of the "rest" energy group.
>> >> Gromacs require table.xvg, even if the rest group is empty.
>> >>
>> >> In the case methanol/ibi the rest group is empty as the whole system
>> >> belongs to the CG energy group (see index.ndx) and the CG-CG
>> >> interaction is covered by a special interaction (grompp.mdp:
>> >> energygrp_table = CG CG).
>> >>
>> >> table.xvg could be useful for a multi-component system, where only one
>> >> interactions is iterated and all the others are covered normal LJ
>> >> interactions.
>> >>
>> >> Christoph
>> >> >
>> >> >
>> >> >
>> >> > Sincerely,
>> >> > Zhicheng
>> >> >
>> >> > 在 2013年7月6日星期六UTC+8上午12时01分18秒，Christoph Junghans写道：
>> >> >>
>> >> >> 2013/7/4 郭志成 <[email protected]>:
>> >> >> > And, how can I generate other intercations table.xvg for gromacs.
>> >> >> I don't really understand your question!
>> >> >>
>> >> >> For the interactions, which you iterate on an no initial guess is
>> >> >> need, you just need the rdf.
>> >> >> For all others (fixed) interaction, you will need a table. Manual
>> >> >> describes this procedure for bonded interaction in some details, for
>> >> >> non-bonded interactions use
>> >> >> CG-CG.pot.new from
>> >> >> $ csg_call --ia-name CG-CG --options settings.xml --ia-type
>> >> >> non-bonded
>> >> >> prepare_single ibi
>> >> >> together with
>> >> >> $ csg_call convert_potential gromacs CG-CG.pot.new table_CG_CG.xvg
>> >> >> to convert CG-CG.pot.new to table_CG_CG.xvg.
>> >> >>
>> >> >> >
>> >> >> > Sincerely,
>> >> >> > Zhicheng
>> >> >> >
>> >> >> > 2013/7/5 郭志成 <[email protected]>
>> >> >> >>
>> >> >> >> Dear Christoph,
>> >> >> >> Thanks for your patient interpretation. I had a misunderstanding
>> >> >> >> about
>> >> >> >> initial guess of non-bonded tabulated potential and other
>> >> >> >> interactions
>> >> >> >> (table.xvg)
>> >> >> >>
>> >> >> >>
>> >> >> >> 2013/7/5 Christoph Junghans <[email protected]>
>> >> >> >>>
>> >> >> >>> 2013/7/4 郭志成 <[email protected]>:
>> >> >> >>> > Hi Christoph,
>> >> >> >>> > Thanks for your reply. Hmm, you're right that it is more
>> >> >> >>> > quicker
>> >> >> >>> > and
>> >> >> >>> > easier
>> >> >> >>> > to run IBI using gromacs than espresso. As mentioned above,
>> >> >> >>> > the
>> >> >> >>> > initial
>> >> >> >>> > tabulated potential for bonded interacation directly can be
>> >> >> >>> > obtained
>> >> >> >>> > using
>> >> >> >>> > csg_boltzmann tool, however, initial guess of tabulated
>> >> >> >>> > potential
>> >> >> >>> > for
>> >> >> >>> > non-bonded interaction can not be generated by csg_boltzamnn
>> >> >> >>> > tool.
>> >> >> >>> > Thus, I
>> >> >> >>> > want to know how to get the initial non-bonded tabulated
>> >> >> >>> > potential
>> >> >> >>> > (csg-tutorial/methanol/ibi/table.xvg) through calculating rdf
>> >> >> >>> > in
>> >> >> >>> > VOTCA.
>> >> >> >>> You don't need to provide an initial guess. VOTCA is doing that
>> >> >> >>> automatically, you could do the same manually using:
>> >> >> >>> $ csg_call --ia-name CG-CG --options settings.xml --ia-type
>> >> >> >>> non-bonded
>> >> >> >>> prepare_single ibi
>> >> >> >>> (change the "CG-CG" to name of your interaction)
>> >> >> >>> This will generate a CG-CG.pot.new, which is the initial guess
>> >> >> >>> calculated from the rdf. You could preprocess that file and give
>> >> >> >>> it
>> >> >> >>> to
>> >> >> >>> Votca as initial guess (CG-CG.pot.in).
>> >> >> >>>
>> >> >> >>> Please don't confuse table.xvg with an initial guess. The
>> >> >> >>> interaction
>> >> >> >>> between the CG beads in methanol/ibi is given as table_CG_CG.xvg
>> >> >> >>> to
>> >> >> >>> gromacs (see the xml settings file). table.xvg is used for all
>> >> >> >>> other
>> >> >> >>> interactions, which are not covered by special tables
>> >> >> >>> (table_X_Y.xvg).
>> >> >> >>> Even if there are no other interactions gromacs will need a
>> >> >> >>> table.xvg
>> >> >> >>> file (this might has changed in gromacs 4.6). None of our
>> >> >> >>> tutorials
>> >> >> >>> actually uses table.xvg, this is just a gromacs specialty.
>> >> >> >>>
>> >> >> >>> Christoph
>> >> >> >>>
>> >> >> >>> >
>> >> >> >>> > Sincerely,
>> >> >> >>> > Zhicheng
>> >> >> >>> > 2013/7/3 Christoph Junghans <[email protected]>
>> >> >> >>> >>
>> >> >> >>> >> 2013/7/3 Zhicheng <[email protected]>:
>> >> >> >>> >> > Hi Christoph,
>> >> >> >>> >> >
>> >> >> >>> >> > I retried the methanol/ibi_espresso tutorial and modified
>> >> >> >>> >> > the
>> >> >> >>> >> > bead
>> >> >> >>> >> > interaction of lennard-jones (inter 0 0 lennard-jones 1.0
>> >> >> >>> >> > 1.0
>> >> >> >>> >> > 1.12246
>> >> >> >>> >> > 0.25
>> >> >> >>> >> > 0.0 ) to replace the tabulated potential (inter 0 0
>> >> >> >>> >> > tabulated
>> >> >> >>> >> > "table_CG_CG.tab" ) in generate.tcl. However, I got the
>> >> >> >>> >> > following
>> >> >> >>> >> > error
>> >> >> >>> >> > message:
>> >> >> >>> >> >
>> >> >> >>> >> >
>> >> >> >>> >> > property not found: table
>> >> >> >>> >> Like I said before the scripts in generate_esp_from_gro are
>> >> >> >>> >> only
>> >> >> >>> >> to
>> >> >> >>> >> generate espresso input data from gromacs input data.
>> >> >> >>> >> (It basically generates conf.esp.gz from conf.gro)
>> >> >> >>> >> If you have gromacs input data already just use gromacs to do
>> >> >> >>> >> IBI,
>> >> >> >>> >> it
>> >> >> >>> >> is easier and quicker!
>> >> >> >>> >>
>> >> >> >>> >> >
>> >> >> >>> >> > In my opinion, it means that the IBI procedure interfaced
>> >> >> >>> >> > with
>> >> >> >>> >> > espresso
>> >> >> >>> >> > only
>> >> >> >>> >> > requires tabulated potential other than another type of
>> >> >> >>> >> > interaction
>> >> >> >>> >> > (such as
>> >> >> >>> >> > lennard-jones). Thus, we have to specify the initial guess
>> >> >> >>> >> > of
>> >> >> >>> >> > tabulated
>> >> >> >>> >> > potential (table_CG_CG.tab). Is that right?
>> >> >> >>> >> No, table_CG_CG.tab is generated from the initial rdf.
>> >> >> >>> >>
>> >> >> >>> >> To run IBI using Espresso just execute run.sh in
>> >> >> >>> >> methanol/ibi_espresso.
>> >> >> >>> >> You might need to modify settings.xml if your Espresso binary
>> >> >> >>> >> is
>> >> >> >>> >> named
>> >> >> >>> >> differently.
>> >> >> >>> >> run.sh execute csg_inverse, which then calls Espresso.
>> >> >> >>> >>
>> >> >> >>> >> Remember to use the updated conf.esp.gz from:
>> >> >> >>> >>
>> >> >> >>> >>
>> >> >> >>> >>
>> >> >> >>> >>
>> >> >> >>> >>
>> >> >> >>> >> <https://csg-tutorials.votca.googlecode.com/hg/methanol/ibi_espresso/conf.esp.gz>
>> >> >> >>> >>
>> >> >> >>> >> Christoph
>> >> >> >>> >> >
>> >> >> >>> >> > Best Regards!
>> >> >> >>> >> >
>> >> >> >>> >> > Zhicheng
>> >> >> >>> >> > 在 2013年6月28日星期五UTC+8下午10时49分24秒，Christoph Junghans写道：
>> >> >> >>> >> >>
>> >> >> >>> >> >> 2013/6/28 Zhicheng <[email protected]>:
>> >> >> >>> >> >> > Dear all:
>> >> >> >>> >> >> >
>> >> >> >>> >> >> > The "hist" ("tab") command within csg_boltzmann tools to
>> >> >> >>> >> >> > generate
>> >> >> >>> >> >> > the
>> >> >> >>> >> >> > distribution function (tabulated potential), after
>> >> >> >>> >> >> > post-processing,
>> >> >> >>> >> >> > can
>> >> >> >>> >> >> > be
>> >> >> >>> >> >> > used to derive bonded interaction ( e.g *.dist.tgt
>> >> >> >>> >> >> > table_a1.tab
>> >> >> >>> >> >> > table_b1.tab
>> >> >> >>> >> >> > in csg-tutorials/propane/ibi_espresso). However, the
>> >> >> >>> >> >> > tabulated
>> >> >> >>> >> >> > potential
>> >> >> >>> >> >> > required by non-bonded interaction can't obtain using
>> >> >> >>> >> >> > this
>> >> >> >>> >> >> > method.
>> >> >> >>> >> >> > Therefore, I want to know how to generate the tabulated
>> >> >> >>> >> >> > potential
>> >> >> >>> >> >> > for
>> >> >> >>> >> >> > non-bonded interaction ( such as table_CG_CG.tab in
>> >> >> >>> >> >> >
>> >> >> >>> >> >> > csg-tutorials/methanol/ibi_espresso/generate_esp_from_gro).
>> >> >> >>> >> >> > I
>> >> >> >>> >> >> > will be
>> >> >> >>> >> >> > appreciative if anyone can help me!
>> >> >> >>> >> >> I think you mixing up two thing here.
>> >> >> >>> >> >>
>> >> >> >>> >> >> 1.) csg_inverse will generate table_CG_CG.tab during IBI
>> >> >> >>> >> >> procedure
>> >> >> >>> >> >> for you, the initial potential is calculated out of the
>> >> >> >>> >> >> rdf
>> >> >> >>> >> >> (rdf_methanol.xvg). So for IBI there is no initial
>> >> >> >>> >> >> table_CG_CG.tab
>> >> >> >>> >> >> needed.
>> >> >> >>> >> >> 2.) The files in generate_esp_from_gro are converts from
>> >> >> >>> >> >> gromacs
>> >> >> >>> >> >> input
>> >> >> >>> >> >> to espresso input (see generate.tcl). This was mainly done
>> >> >> >>> >> >> to
>> >> >> >>> >> >> allow
>> >> >> >>> >> >> user to compare the behavior of different simulation
>> >> >> >>> >> >> back-ends
>> >> >> >>> >> >> (espresso vs gromacs). In a real world study, there are
>> >> >> >>> >> >> very
>> >> >> >>> >> >> few
>> >> >> >>> >> >> cases
>> >> >> >>> >> >> where one would convert gromacs input files to espresso
>> >> >> >>> >> >> instead
>> >> >> >>> >> >> of
>> >> >> >>> >> >> simply using gromacs directly.
>> >> >> >>> >> >>
>> >> >> >>> >> >> Christoph
>> >> >> >>> >> >> >
>> >> >> >>> >> >> > Best Regards!
>> >> >> >>> >> >> >
>> >> >> >>> >> >> > Zhicheng
>> >> >> >>> >> >> >
>> >> >> >>> >> >> > --
>> >> >> >>> >> >> > 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.
>> >> >> >>> >> >> >
>> >> >> >>> >> >> >
>> >> >> >>> >> >>
>> >> >> >>> >> >>
>> >> >> >>> >> >>
>> >> >> >>> >> >> --
>> >> >> >>> >> >> 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.
>> >> >> >>> >> >
>> >> >> >>> >> >
>> >> >> >>> >>
>> >> >> >>> >>
>> >> >> >>> >>
>> >> >> >>> >> --
>> >> >> >>> >> Christoph Junghans
>> >> >> >>> >> Web: http://www.compphys.de
>> >> >> >>> >>
>> >> >> >>> >> --
>> >> >> >>> >> You received this message because you are subscribed to a
>> >> >> >>> >> topic
>> >> >> >>> >> in
>> >> >> >>> >> the
>> >> >> >>> >> Google Groups "votca" group.
>> >> >> >>> >> To unsubscribe from this topic, visit
>> >> >> >>> >>
>> >> >> >>> >> https://groups.google.com/d/topic/votca/1ab9M7DmThk/unsubscribe.
>> >> >> >>> >> To unsubscribe from this group and all its topics, 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].
>> >> >> >>> > 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.
>> >> >> >>> >
>> >> >> >>> >
>> >> >> >>>
>> >> >> >>>
>> >> >> >>>
>> >> >> >>> --
>> >> >> >>> Christoph Junghans
>> >> >> >>> Web: http://www.compphys.de
>> >> >> >>>
>> >> >> >>> --
>> >> >> >>> You received this message because you are subscribed to a topic
>> >> >> >>> in
>> >> >> >>> the
>> >> >> >>> Google Groups "votca" group.
>> >> >> >>> To unsubscribe from this topic, visit
>> >> >> >>> https://groups.google.com/d/topic/votca/1ab9M7DmThk/unsubscribe.
>> >> >> >>> To unsubscribe from this group and all its topics, 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].
>> >> >> > 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.
>> >> >> >
>> >> >> >
>> >> >>
>> >> >>
>> >> >>
>> >> >> --
>> >> >> 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.
>> >> >
>> >> >
>> >>
>> >>
>> >>
>> >> --
>> >> Christoph Junghans
>> >> Web: http://www.compphys.de
>> >>
>> >> --
>> >> You received this message because you are subscribed to a topic in the
>> >> Google Groups "votca" group.
>> >> To unsubscribe from this topic, visit
>> >> https://groups.google.com/d/topic/votca/1ab9M7DmThk/unsubscribe.
>> >> To unsubscribe from this group and all its topics, 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].
>> > 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.
>> >
>> >
>>
>>
>>
>> --
>> 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.
>
>
--
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.
