[R] genotypes simulation

[Neil Shephard]

Claire_6700 wrote:
> 
> Hello,
> 
> I am having really hard time finding a good article about simulating
> genotypes of cases and controls at a disease locus using R.
> 
> if you guys can point me or guide me where i can find more information, it
> will be helpful.
> 
> 

The popgen() package allows the simulation of genotype data under a
coalescent model (via the treesim() function) or Multinomial-Drichlet model
(via the simMD() function).  These won't quite simulate case-control data,
but can no doubt be tweaked to get two sets of data (cases and controls)
with the desired allele frequencies, LD etc.

There is also the rmetasim() package which interfaces R with metasim.

Alternatively if your after validation of p-values derived from genotypic
tests of association via simulation there is also the Direction Simulation
Approach (DSA) which is implemented in R
(http://www.mpipsykl.mpg.de/pages/english/research/mueller_downloads.htm).

There is also an R plugin for plink
(http://pngu.mgh.harvard.edu/~purcell/plink/) which performs simulation of
unlinked loci that are in linkage equilibrium (although you may specifically
want to model linkage disequilibrium).

And there is an R-plugin for PBAT which is for Population Based Association
Tests (http://people.fas.harvard.edu/~tjhoffm/pbatR.html).

For other R-genetics information you may find the task-view at
http://cran.r-project.org/web/views/Genetics.html useful as well as
http://rgenetics.org/

There are also a host of other non-R options available though.  Many under
the coalescent model (including one of the first by Richard Hudson, the ms
program) are linked from
http://en.wikipedia.org/wiki/Coalescent_theory#Software and additional
software for genetic analysis (including simulation in some instances)  at
http://linkage.rockefeller.edu/soft/ .  The vast majority will simply dump
output to text files, so its straight-forward to call them from within R via
a system call and then read the results into R.

Neil
-- 
View this message in context: http://www.nabble.com/genotypes-simulation-tp17065607p17066332.html
Sent from the R help mailing list archive at Nabble.com.