Re: Spinangle with refgroup : Warning: discontinuous rotation!

From: nicolas martin (nicolasmartin973_at_gmail.com)
Date: Thu Apr 10 2014 - 08:30:44 CDT

Dear Giacomo, dear all,

I'm sorry for the misent messages I got troubles to do that simple thing..

You are right maybe I should have explained what was my point of making a
comparison between both spinangle and dihedral which are too different
things. I know how to manage the colvar dihedral and I find it pretty
simple to use but I cannot say that about spinangle. I figure out that in
special cases, if I understood right reading the UG, spinangle and dihedral
might give the same result (maybe just with a shift due to the reference
structure).

To illustrate what I've just said :

If C1, C2 and C3 remain fixed by for example using the keyword fixedAtoms
or applying an absolute restrain on them, defining the axis of rotation in
the colvar spinangle as C2-C3 bond. and using the C4 and its H to measure
the spinangle, I thought that I could correlate dihedral and spinangle. But
I was obviously mistaken.

My final goal is to compute the PMF along a coordinate reaction e.i. the
twisting angle between the trans-membrane and extracellular domain of a
receptor. And doing so with and without a ligand. What do you think would
be the best approach to do so ? I have been watching as well to the
possibility of SMD. What is you opinion about applying it to my problem ?

Thank you again for your help and reactivity, bests,

NM

2014-04-09 18:18 GMT+02:00 Giacomo Fiorin <giacomo.fiorin_at_gmail.com>:

> Hello Nicolas, here are some suggestions:
> - please send all your emails to the mailing list, not only part of them;
> - butane is a tiny molecule, but still fairly flexible: please try a
> really rigid one such as benzene or phenol instead of setting constraints;
> - when you constrain/restrain the positions of heavy atoms, the two
> terminal methyl groups can still rotate;
> - a dihedral angle and a spinangle are two different quantities: your
> setup is only for the latter, if you wish to define the dihedral
> C1-C2-C3-C4, you should do so by using a "dihedral" function.
>
> Regards,
> Giacomo
>
>
>
> On Wed, Apr 9, 2014 at 11:40 AM, nicolas martin <
> nicolasmartin973_at_gmail.com> wrote:
>
>> Dear Giacomo,
>>
>> After reading your message I went back to my first idea which was
>> developing my method on a way smaller molecule : butane. As you mentioned I
>> tried to solve the problem of RMSD on both group by restraining them. I
>> still have those Warnings and actually cannot align my reference group in
>> VMD. In another hand, I can see clearly the spinangle colvar going from 0
>> to 40 as asked but the dihedral angle itself is not moving... I think this
>> might be due to a failure of the alignment.
>>
>> Here is the input file :
>>
>> colvar {
>> name spinangle
>> spinangle {
>> axis (-1.408,-0.537,-0.317) # axis C2 C3
>>
>> atoms {
>> atomNumbers 11 12 13 14 #..moving group
>> centerReference on
>> rotateReference on
>> refPositionsGroup { #..reference group
>> atomNumbers 1 5 8
>>
>> }
>> refPositionsFile input/ref.pdb
>> refPositionsCol O
>> refPositionsColValue 1.00
>> }
>>
>> refPositionsFile input/ref.pdb
>>
>> }
>> }
>> harmonic {
>> name harmonic
>> colvars spinangle
>> forceConstant 0.0060923
>> centers 0
>> targetCenters 40
>> targetNumSteps 1000
>> targetNumStages 11
>> }
>> colvar {
>> name rmsd
>> rmsd {
>> atoms { atomNumbers 11 12 13 14 }
>> refPositionsFile input/ref.pdb
>> }
>> }
>> harmonic {
>> name rmsd_hamonic
>> colvars rmsd
>> forceConstant 100
>> centers 0
>> }
>>
>
>

This archive was generated by hypermail 2.1.6 : Wed Dec 31 2014 - 23:22:19 CST