Temperature coupling

You shouldn't couple just a handful of atoms seperately to a heat bath. Or you should at least not be surprised to see large fluctuations when you do. If you only have a few atoms, the temperature, depending on the average speed, will fluctuate a lot. When you have more atoms, it averages out. Moreover, coupling the temperature to a fair amount of atoms will only involve small changes on the speeds of the atoms, whereas coupling to  a few atoms may give considerable alterations and thus cause artefacts in  your simulation.  The best thing to do is to merge the ions and solvent groups and couple  this to the heat bath.  Cheers,  Tsjerk   --  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -- :) -- :)  Tsjerk A. Wassenaar, M.Sc. -- :)  Molecular Dynamics Group -- :)  Dept. of Biophysical Chemistry -- :)  University of Groningen -- :)  Nijenborgh 4 -- :)  9747 AG Groningen -- :)  The Netherlands -- :) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Metabolomics

What is metabolomics?
The idea of ‘metabolomics’ has been coined and developed in the last decade to
comprehensively study metabolism under genetic and environmental perturbations .However, the first papers involving metabolite profiling techniques were
published well over 30 years ago, with the aim, at that time, of rapid medical diagnostics
. The underlying idea behind the use of metabolomics in plant biology
today is to detect metabolic effects of genetic or environmental perturbation which
may only distantly relate to known or presumed primary (enzymatic) alterations.
Metabolomics, therefore, seeks to detect ‘unexpected’ events on a comprehensive
scale, and it widely acknowledges the presence of novel metabolites with unknown
chemical structure or biological function.
In this respect, it differs from classical control theory that has been applied more
frequently to select well-known pathways or regulatory circuits with the objective
to understand these pathways in a mathematical manner using well-defined models
and assumptions. Usually, mathematical control models need to be supported by
high level metabolite measurements such as flux data. Although some efforts have
been reported to derive larger metabolic models from isotope calculations of protein
hydrolysates, we are still far away from reaching the goal universal and global
‘fluxome’analysis, especially with regard to plant research. Metabolomics does
not try to reach this goal. Its use in studying metabolism has so far been more of an
observatory and confirmatory role. It aims less at directly deriving insights into the
cellular organization of metabolism. Due to its power to detect broad classes of
metabolites, including unknowns, metabolomics is best used for studying system
properties (such as networks) and changes (control) of metabolite levels in disparate
parts of metabolism.