Re: Rotate two linked proteins against each other and calculate energy profile

From: Bjoern Olausson (namdlist_at_googlemail.com)
Date: Mon Mar 28 2011 - 07:49:23 CDT

Hi Ajasja,

you made a good point here. The homodimer has 2423 Atoms so if I would select
the entire protein this would split something like A) 1100, B)100, C) 100,
D)1100
Choosing only the C-Alpha, I'll probably should include the hydrogens on the
C-Alphas too since I am using shake, would shrink the selection to
A) 146, B) 6, C) 8, D) 148

I would have taken a detour if you wouldn't have mentioned ABF! Thanks a lot
for this comment!

So after playing around with NAMD CVS, GBIS, COLVARS, and ABF I could verify
that my setup works (I just calculated the free energy profile for ethane and
I could reproduce a nice sine like profile for -180 to 180 degree rotation
with a peak at around 2,7kcal for the eclipsed conformation). Still I have
some questions how ABF works:

Why can't I reproduce the above when choosing "lowerBoundary 0" and
"upperBoundary 360" instead of -180 and 180? -180 and 180 works fine, same
setup but 0 and 360 results in a higher energy at 120 degree (4.2 kcal) and
produces a flat line at 2 kcal from 180 to 360 degrees.
See the following graphics:
ethane_0_to_360.png
http://ompldr.org/vODAzbw/ethane_0_to_360.png
ethane_-180_to_180.png
http://ompldr.org/vODAzcA/ethane_-180_to_180.png

If a bin is poorly sampled, one can interpret this as a highly unfavored
conformation or the applied force was not sufficient to push the structure
into this conformation and the process might not be reversible. Is this
assumption right (assuming that the sampling time was sufficient)?

For the reversibility part:
The ABF tutorial is talking about plotting Xi (Reaction Coordinate) against
the time. Is there a built-in keyword which outputs Xi as a function of time
or can I make NAMD do so in some way. Okay, it is pretty easy with VMD, but if
I could do it alongside with the NAMD calculation itself, this would be very
nifty and would save some post processing ;-)

Thanks again for your help!
I appreciate it very much.

Cheers,
Bjoern

On Monday 21 March 2011 09:43:55 Ajasja Ljubetič wrote:
> Hi Bjoern,
>
> The analogy with ethane is reasonable. So is the definition of the points.
> A little though has to be given about which atoms of the subunit will be
> included. If you look at the recommendations for colvar
> definitions<http://www.ks.uiuc.edu/Research/namd/2.7/ug/node49.html#SECTION
> 000122310000000000000> you
> will note that a lot of atoms in the atom group may hurt scaling and
> performance of NAMD. I do not know how big is each subunit. Perhaps is you
> include just the C-alpha atoms...?
>
> By using the adaptive biasing force
> calculations<http://www.ks.uiuc.edu/Research/namd/2.7/ug/node50.html#SECTIO
> N000123100000000000000> you
> do not have to apply any additional forces. Just include your colvar in the
> abf block and you're good to go.
>
> Best regards,
> Ajasja
>
> On Sun, Mar 20, 2011 at 17:03, Bjoern Olausson wrote:
> > Thanks again for your suggestion Ajasja, I am still planing the setup
> > and I finally decided to go with "colvars" and "SMD".
> >
> > I am mulling over your "colvars" and Dihedral suggestion.
> > Since it is the first time I am using "colvars" I am somewhat
> > uncertain how I should group the atoms.
> >
> > My current Idea to define a dihedral would be the following:
> > Group 1: Monomer A
> > Group 2: First half of the linker
> > Group 3: Second half of the linker
> > Group 4: Monomer B
> >
> > Does this make sens to you?
> >
> > To come back to the Ethan molecule, Group 1 would resemble a Proton on
> > C1, Group 2 would be C1 itself, Group 3 would be C2 and Group 4 would
> > be a Proton on C2.
> >
> > Is this analogy correct?
> >
> > Finally I would apply a force to Monomer A (Group 1) to rotate it
> > (maybe I have to constrain Monomer B, but we will see)
> >
> > Does this sound reasonable?
> >
> > Thanks a lost for your suggestions,
> > Bjoern
> >
> > On Tue, Mar 15, 2011 at 22:56, Ajasja Ljubetič
> >
> > <ajasja.ljubetic_at_gmail.com> wrote:
> > > Hi,
> > > Perhaps the colvars module would be of interest to you as well. You
> > > could define a dihedral colvar between the two subunits and obtain the
> > > free
> >
> > enrgy
> >
> > > profile of the rotation.
> > > Best regards,
> > > Ajasja
> > >
> > >
> > > On Tue, Mar 15, 2011 at 20:20, Bjoern Olausson
> > > <namdlist_at_googlemail.com>
> > >
> > > wrote:
> > >> Hi NAMD users,
> > >>
> > >> I have a homo dimer which is linked via a short peptide linker. The
> > >> two monomers are supposed to rotate more or less free against each
> > >> other.
> > >>
> > >> I would like to follow the energy profile while one monomer is rotated
> > >> 180 degrees to see if there is on favorite position. Something like
> > >> the energy profile when rotating the dihedral angle of an Ethane
> > >> molecule.
> > >>
> > >> I already have a explicite solvent, all-atom simulation running for
> > >> this system.
> > >>
> > >> Now my qestions:
> > >> a) Would I use FEP for the energy profile (Use coordinates from my
> > >> simulation as start structure and as target structure take coordinates
> > >> from a SMD run where I rotate one monomer by 180 degrees)? I guess
> > >> not. Since I can't tell FEP where to go.
> > >>
> > >> b) Instead of using FEP, is it possible to just run a SMD where I
> > >> rotate (rotConstraints) one monomer by 180 degree and calculate the
> > >> forces from the SMD output (like in the NAMD tutorial "Force Analysis
> > >> for Constant Velocity Pulling"
> >
> > http://www.ks.uiuc.edu/Training/Tutorials/namd/namd-tutorial-unix-html/no
> > de19.html#SECTION00064100000000000000 )?
> >
> > >> c) Is it possible to rotate a molecule using tcl forces instead of
> > >> SMD?
> > >>
> > >> Thanks for you help.
> > >>
> > >> Cheers,
> > >> Bjoern

-- 
Bjoern Olausson
Martin-Luther-Universität Halle-Wittenberg 
Fachbereich Biochemie/Biotechnologie
Kurt-Mothes-Str. 3
06120 Halle/Saale
Phone: +49-345-55-24942

This archive was generated by hypermail 2.1.6 : Mon Dec 31 2012 - 23:20:00 CST