Uses the MaCS software to produce founder haplotypes MaCSAlphaSimR.
Usage
runMacs(
nInd,
nChr = 1,
segSites = NULL,
inbred = FALSE,
species = "GENERIC",
split = NULL,
ploidy = 2L,
manualCommand = NULL,
manualGenLen = NULL,
nThreads = NULL
)Arguments
- nInd
number of individuals to simulate
- nChr
number of chromosomes to simulate
- segSites
number of segregating sites to keep per chromosome. A value of NULL results in all sites being retained.
- inbred
should founder individuals be inbred
- species
species history to simulate. See details.
- split
an optional historic population split in terms of generations ago.
- ploidy
ploidy level of organism
- manualCommand
user provided MaCS options. For advanced users only.
- manualGenLen
user provided genetic length. This must be supplied if using manualCommand. If not using manualCommand, this value will replace the predefined genetic length for the species. However, this the genetic length is only used by AlphaSimR and is not passed to MaCS, so MaCS still uses the predefined genetic length. For advanced users only.
- nThreads
if OpenMP is available, this will allow for simulating chromosomes in parallel. If the value is NULL, the number of threads is automatically detected.
Value
an object of MapPop-class
Details
There are currently three species histories available: GENERIC, CATTLE, WHEAT, and MAIZE.
The GENERIC history is meant to be a reasonable all-purpose choice. It runs quickly and
models a population with an effective populations size that has gone through several historic
bottlenecks. This species history is used as the default arguments in the runMacs2
function, so the user should examine this function for the details of how the species is modeled.
The CATTLE history is based off of real genome sequence data cattleAlphaSimR.
The WHEAT gaynor_2017AlphaSimR and MAIZE hickey_2014AlphaSimR histories have been included due to their use in previous simulations. However, it should be noted that neither faithfully simulates its respective species. This is apparent by the low number of segregating sites simulated by each history relative to their real-world analogs. Adjusting these histories to better represent their real-world analogs would result in a drastic increase to runtime.