Re: * CGtool force field *

From: Axel Kohlmeyer (
Date: Sat Oct 22 2011 - 14:36:38 CDT

On Sat, Oct 22, 2011 at 3:11 PM, HAMID - <> wrote:
> Dear Dr. Kohlmeyer,
> Thank you for your reply.
> You are right about that I did not find any answer.
> I should mentioned in my previous post that I checked all atom FF,
> like charmm, for my issue but all atom MD was time consuming for me.
> Thus I explored another method such as shape based and residue based
> coarse graining methods. I decided to study the mechanical properties
> of histone complex and initiation factor complex so I thought my
> problem may have a solution via RBCG.

sorry, but that is not a good argument. what would you
think of somebody who tells you "i just use 3 to
represent 'pi', since i cannot afford a pocket calculator"?
the only place where you can do such an
approximation are cases where a 5% error in 'pi'
has no significant impact on the outcome.

..of course residue based coarse graining is a much
more crude approximation than using 3 instead of 'pi'.

> Maybe I did not express my idea about protein-protein complex clearly.
> I decide to study the "sliding" of histone complex and initiation
> factor complex on each other. ( I am informed of that a classical MD
> package like NAMD is not able to reproduce QM calculations to study
> bond breaking.)

have you done this with the available model?
is there any indication outside of your hunch
that the missing polarizability may affect your
results. if that is the case, i would worry a lot
about coarse graining in the first place, too.

> About the cg water, I know the RBCG water bead represent a cluster of
> water which contains four water molecules. In my first post about RBCG

and which are neutral.

> water I asked about the polarizble CG water model like that the
> martini polarizable water provides. I have tried several times to
> translate martini polarizble water top and par file to CHARMM
> counterpart but I could not reproduce a valid charmm like CG

no surprise there. you are doing the equivalent
of trying to fit a round peg into a square hole.

> polarizable water model yet. I hope I design a polarizable CG water
> model for using in NAMD. So I ask every body to help me in this
> regard.

apparently, you are the only one vocal about wanting this,
so don't get your hopes up that through some magic intervention
this will suddenly happen. one of the major rules of open source
software is: if you want it, you gotta do it.

why don't you start with using gromacs instead. if you want
to implement this in NAMD (which will likely require some
programming), then you'll need a reference to compare to.

> Hope I have clarified that I did not complain about not getting any
> answer. I have learned that I should study well and work hard to find
> an answer. If I do not find reliable answer then I ask from experts
> like you.

if there is nothing that says that the feature you are looking for
is implemented than changes are extremely high that, in fact,
it doesn't exist. it doesn't take an expert to see this, just
common sense.

> Excuse me if my questions seem not advanced. I am a beginner like any
> expert was.

> I am still looking forward to hearing comments.

well, now you've got some. satisfied?


> Best,
> On 10/22/11, Axel Kohlmeyer <> wrote:
>> On Sat, Oct 22, 2011 at 1:00 PM, HAMID - <> wrote:
>>> Dear all,
>>> I am looking forward to every one answer my previous post about RBCG
>>> water.
>> dear hamid,
>> if you don't get an answer, then usually for a reason.
>> perhaps nobody thought that your question was
>> worth answering, perhaps people thought the answer
>> is too obvious, perhaps people didn't understand
>> the question, perhaps nobody knows an answer.
>>  the remedy for each of these is that *you* have
>> to think more about your problem, work harder
>> and come up with a better/simpler question instead
>> of complaining about not getting any. complaining
>> only makes it _less_ likely to get a qualified answer.
>>> I check the RBCG water bead top and parameter file which implemented
>>> in CGtool folder of VMD.
>>> Again I see the water beads in RBCG have zero charge thus, such water
>> pop quiz: what does a water bead in this model
>> represent and what is the total charge of that?
>>> model does not able to reproduce electrostatic interactions and I
>>> think it is a disaster for my simulations.
>> well, like the saying goes: if the cap don't fit, don't wear it.
>>> Does any one introduce me a method to compensate the absence of water
>>> charge effects in my RBCG simulations (simulation of a huge protein
>>> complex - protein interactions)?
>> this is a question that is best answered by studying
>> the literature. it looks to me that you are very confused
>> about what a coarse grain model can do and what not.
>> it is pointless to study something that cannot be properly
>> represented with a model you chose. you wouldn't want
>> to use a classical force field to study bond breaking,
>> or would you?
>>> I would really appreciate any guide.
>> my advice: get busy learning the method and its background
>> before actually starting something that may be a waste of time.
>> axel.
>>> Best.
>>> ----
>>> H. Hadi
>>> PhD student of Biophysics,
>>> UT, Iran
>> --
>> Dr. Axel Kohlmeyer
>> Institute for Computational Molecular Science
>> Temple University, Philadelphia PA, USA.
> --
> ----
> H. Hadi
> PhD student of Biophysics,
> UT, Iran

Dr. Axel Kohlmeyer
Institute for Computational Molecular Science
Temple University, Philadelphia PA, USA.

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:57:51 CST