Re: Helix tilt

From: Thomas C Bishop (bishop_at_tulane.edu)
Date: Tue Jul 13 2010 - 01:08:45 CDT

Just a note of caution about fitting an axis to a helix...

you must be very careful about defining the beginning and ending points as
this will tend to weight the helix axis towards one direction or the other. I
tried the simple moments inertia approach a long time ago and found that even
if considering 3 to 5 turns of the helix the "begin-end" effect can cause a
significant deviation of the axis.

It will be important for you to demonstrate how big/small these affects are.
This is easily done by measuring the angle between various definitions of
helix axis as applied to ONE helix.

Then you can compare two helices w/ some idea of the error bars.
If the helix is not straight there will also be problems

Your best bet is to find something based on proper mathematical
analysis/optimization techniques. HELANAL seems to take this approach so I'd
be very interested in how the choice of helix "beg-end" residues affects
HELANAL's idea of the helix axis.

Tom

On Tuesday 06 July 2010 12:27:50 pm Eduard Schreiner wrote:
> Hi guys,
>
> a quite simple solution in terms of post-processing of a trajectory
> would be to calculate the inertia tensor for both helices and then
> calculate the angle between the axes corresponding to the smallest
> moment of inertia; this can be done in VMD.
> For an ideal helix, this axis should correspond to the helix axis.
> If the helix is bent, this correspondence does not hold anymore.
> However, then the relative angle between two helices may be not the
> proper descriptor anyway.
>
>
> eddi
>
> On Tue, Jul 6, 2010 at 11:48 AM, Nicholas M Glykos <glykos_at_mbg.duth.gr>
wrote:
> >> how would one calculate the tilt between two helices along a
> >> simulation in namd. I would like to know if there are any analysis
> >> scripts written for it or tool or something like that.
> >
> > You are possibly posting to the wrong list. Having said that, calculating
> > crossing angles for associating helices is not as straightforward as you
> > may think. You would normally have to calculate the local helix axis for
> > each helix, and then work your way from there. Once upon a time, there
> > was a program called HELANAL (written in FORTRAN77(?), that's why I am
> > using capital letters) which could do the trick (and much more). I bet
> > there are more recent and automated programs available these days, but a
> > quick search showed that HELANAL is still available. Needless to say that
> > you probably can't do the calculation directly with your DCD file (save
> > PDB files from the DCD and work with them).
> >
> > Nicholas
> >
> >
> > --
> >
> >
> > Dr Nicholas M. Glykos, Department of Molecular Biology
> > and Genetics, Democritus University of Thrace, University Campus,
> > Dragana, 68100 Alexandroupolis, Greece, Tel/Fax (office) +302551030620,
> > Ext.77620, Tel (lab) +302551030615, http://utopia.duth.gr/~glykos/
>

-- 
**************************
Thomas C Bishop
Tel 504-862-3370 
Fax 504-862-8392
http://dna.ccs.tulane.edu
**************************

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:23:05 CST