[MMTK] about the spring constant

Konrad Hinsen research at khinsen.fastmail.net
Mon Aug 26 12:28:16 UTC 2013

子茗 writes:

 > when use the GNM to calculate the dynamics of a protein, there is a
 > variable to be set,the spring constant.

 > However, I have no idea about it, how to set it?  some papers have
 > referred to the least square fitting? how is the procedure?

For most applications, the basic rule of elastic network models is: if
multiplying all your force constants by some constants makes a
difference for your results, then your results are probably wrong

Elastic network models describe an effective potential energy well
around a stable conformation of a protein. The overall scale of that
well should correspond to the amplitudes of the thermal fluctuations
in the protein. The big problem is that there is no reliable source of
information for obtaining such numbers. No experimental technique
provides a sufficiently direct access, and MD simulations suffer from
insufficient sampling.

When you read about people least-squares fitting the scale factor,
they just try to hide the problem. They fit their ENM results to
whatever quantity they compare to. Which is OK as long as it is clear
that this is just a way to get rid of a problem for which there is
currently no solution.

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
ORCID: http://orcid.org/0000-0003-0330-9428
Twitter: @khinsen

More information about the mmtk mailing list