[MMTK] Value Error: cyclic bond structure

Konrad Hinsen research at khinsen.fastmail.net
Wed Oct 28 17:10:36 UTC 2009


On 27.10.2009, at 19:29, tezcan at ecs.umass.edu wrote:

> While trying to read phi/psi angles of some residues, I get the  
> error "Value Error: cyclic bond structure". I am wondering why this  
> might be happening, and if there is any way to overcome it. I ran in  
> to this error in many residues of 1CBN and 2ERL. As an example:

If all you want is calculate phi-psi angles for a given structure, use

	universe.protein[0][10].phiPsi()

This returns both phi and psi and won't complain about cyclic bond  
structures.

The method phiAngle() returns a dihedral angle object that allows to  
calculate AND modify the phi angle. Modifying a dihedral implies by  
definition that all other internal coordinates remain unchanged. It is  
done by rotating the atoms of the molecule on one side of the dihedral  
axis relative to the atoms on the other side. This requires that the  
two sides are clearly defined, which in turn requires the absence of  
cycles in the bond structure. In fact, when there is a cyclic bond  
structure, it is impossible to modify one dihedral angle without  
modifying any other internal coordinate.

In proteins, the most frequent cause for cyclic bond structures is a  
disulfide bond. I suspect your two proteins contain at least one.

Konrad.
--
---------------------------------------------------------------------
Konrad Hinsen
Centre de Biophysique Moléculaire, CNRS Orléans
Synchrotron Soleil - Division Expériences
Saint Aubin - BP 48
91192 Gif sur Yvette Cedex, France
Tel. +33-1 69 35 97 15
E-Mail: research at khinsen dot fastmail dot net
---------------------------------------------------------------------










More information about the mmtk mailing list