SPASIBA: Continuous Spatial Assignment

Description Usage Arguments Value Author(s) Examples

Simulate data from the prior/likelihood model assumed in SPASIBA

SPASIBA.sim(
  ploidy = 2,
  n.loc = 10,
  n.quant = 0,
  sphere = FALSE,
  size.pop.ref,
  coloc.ref = FALSE,
  n.unknown,
  nx.grid,
  ny.grid,
  win,
  kappa.geno = NULL,
  nu.geno = 1,
  sigma2.geno = NULL,
  kappa.quant = NULL,
  nu.quant = 1,
  sigma2.gf.quant = NULL,
  marginal.quant = "lnorm",
  tau.noise.quant = NULL,
  store.gf = FALSE,
  obsolete = FALSE
)

`ploidy`	1 or 2
`n.loc`	integer nuber of loci
`n.quant`	number of quantitative covariates
`size.pop.ref`	Matrix with one row per population and one column per locus giving haploid population sample size
`coloc.ref`	TRUE / FALSE indicating whether colocated samples are allowed
`n.unknown`	Number of individuls with unknown spatial coordinates
`nx.grid`	Number of pixels horizontally
`ny.grid`	Number of pixels vertically
`win`	c(x.min,x.max,y.min,y.max)
`kappa.geno`	spatial scale parameter, hidden Gaussian RF, genotypes
`nu.geno`	spatial smoothing parameter, hidden Gaussian RF, genotypes
`sigma2.geno`	marginal variance, hidden Gaussian RF, genotypes
`kappa.quant`	spatial scale parameter, quant variables
`nu.quant`	spatial smoothing parameter, quant variables
`sigma2.gf.quant`	marginal variance, quant variables
`marginal.quant`	distribution of quantitative variables ('norm' or 'lnorm')
`tau.noise.quant`	precision of distribution of quantitative variables
`store.gf`	TRUE / FALSE indicating whether Gaussian random field values used to simulated allele frequencies should be stored
`obsolete`	TRUE / FALSE indicating whether random field simulations should be done with the obsolete function GaussRF or the more recent function RFsimulate
`flat`	domai

A list with components

sim.method: The simulation method. The function currently implements the "geostat" model
sphere: A logical indicating whether coordinates are handled as (Lat,Lon) on the sphere or Euclidean coordinates in a 2D flat domain
geno.ref: A matrix containing the genetic data (individual or
population allele counts) of the reference samples. This matrix has one line per population (or individual) and one column per locus
coord.ref: A matrix containing the geographic coordinates of
reference populations. This matrix has one line per population (or individual) and two columns
size.pop.ref: A matrix containing the haploid population size of reference populations. This matrix has one line per population (or individual) and one column per locus
geno.unknown: Indivudual genotypes of individuals of unknown geographic origin
true.coord.unknown: True geographic coordinates of individuals assumed here to be of unknown geographic origin
ploidy: 1 or 2
kappa.geno: Scale parameter used in the simulation of the underlying Matern field
nu.geno: Smoothing parameter used in the simulation of the underlying Matern field
sigma2.geno: Variance parameter used in the simulation of the underlying Matern field
coord.grid=coord: Coordinates of the regular grid at which simulation of Gaussian fields and allele frequencies is performed

Gilles Guillot

dat <- SPASIBA.sim(ploidy=2,
                   n.loc=3, # nb of loci
                   sphere=FALSE,
                   size.pop.ref=rep(2,30), # haploid pop size at each site
                   n.unknown=20, # nb indiv with unknown origin
                   nx.grid=100, # nb hor. pixels on grid 
                   ny.grid=100, # nb vert.  pixels on grid 
                   win=c(0,10,40,50),
                   kappa.geno=1/10,# spatial scale parameter
                   nu.geno=1,# spatial smoothing parameter
                   sigma2.geno=1# marginal variance of hidden Gaussian RF
                   )
                   summary(dat)