[R] rmetasim: a population genetic simulation environment

Allan Strand stranda at cofc.edu
Tue Jan 6 03:32:54 CET 2004

Hi all,

My student, James Niehaus, and I have been working on an
individual-based population genetic simulation package in R.
Currently, the package is still rough, but useful enough that it may
be of interest to population/ecological geneticists.

The best description of the basic model (implemented in C++) can be
found in:

Allan E. Strand. Metasim 1.0: an individual-based environment for
simulating population genetics of complex population
dynamics. Mol. Ecol. Notes, 2:373-376, 2002.

My idea was to produce an extremely flexible engine that could
simulate genotypic data that result from most any demographic
scenario.  These data can be used as null distributions to compare to
observed datasets.  Results of simulations can be exported to a
variety of canned population genetic analysis programs, and rmetasim
implements a few rudimentary analyses (e.g Weir&Cockerhams theta,
mismatch distributions, and assignment tests) as well.  Several
example sessions are included in pdf files found in the rmetasim/doc

rmetasim is mostly a wrapper for the C++ engine described in the paper
cited above.  Because it simulates individuals directly, rmetasim is
not terribly fast, even though the majority of the processing occurs
in compiled code.  Nevertheless, I have found it useful, and running
multiple simulations from the same starting conditions seems to work
in a cluster environment using Rmpi.

We are actively working on this package, and would appreciate feedback
so that we can improve its quality.

The source distribution can be found at:

It should compile on linux boxes (that have R installed), though doing
so takes a while.

Binary distribution for Mac OS X:

Binary distribution for Windows:

Allan Strand,   Biology    http://linum.cofc.edu
College of Charleston      Ph. (843) 953-9189
Charleston, SC 29424       Fax (843) 953-9199

More information about the R-help mailing list