microsimr provides functions for simulating microsatellite data under a coalescent using Hudson's ms program. The user is able to analyze the simulated data directly within R. There is also the option of outputting the simulated data in a format suitable for Migrate-N. Functions to output simulated data in other formats can be added upon request (e.g., IMa2 or LAMARC). Currently, microsimr only simulates microsatellite data under a symmetrical strict stepwise mutation model, and a two-phase mutation model with symmetrical geometric distribution. Other mutation models can be considered if there is interest.
To install microsimr you need Hadley Wickam's devtools:
install.packages("devtools")
You can then install microsimr by:
devtools::install_github("andersgs/microsimr")
If you want the vignettes to be installed, you should use:
devtools::install_github("andersgs/microsimr", build_vignettes = TRUE)
But, this will take a little longer to install.
To simulate 100 loci at theta = 5 (per locus mutation rate scaled by the effective population size --- 4Nemu) genotyped across 20 individuals, with 10 individuals sampled from each of 2 populations exchanging migrants at the rate of 15 migrants per generation (4Nem = 15):
simd_data <- microsimr::sim_microsats(theta = 5,
n_ind = c(10,10),
n_loc = 100,
n_pop = 2,
ms_options = "-I 2 20 20 15")
The data can then be saved to a Migrate-N formatted input file thus:
microsimr::save_microsimr(genotypes = simd_data,
filename = "infile",
n_ind = c(10,10),
n_pop = 2,
n_loc = 100,
format = 'migrate-n')