sim.mpcross: Simulate data from multi-parent designs

Description

Data is simulated according to a pedigree, map and QTL model

Usage

sim.mpcross(map, pedigree, qtl = NULL, vare = 1, error.prob = 0, missing.prob = 0, full.prob = 0, keep.qtlgeno = TRUE, transpos = 1, transval = 0, map.function = c("haldane", "kosambi"), seed = 1, fg = NULL, founderld = FALSE)

Arguments

map

Linkage map with which to generate data. See sim.map

pedigree

Pedigree for a multi-parent cross. Can be generated using sim.mpped

qtl

QTL model, defined by a matrix with one row per QTL and 6 columns: the chromosome of the QTL, the position in cM on that chromosome, and the four founder effects

vare

Phenotypic error variance

error.prob

Probability of genotyping errors - data will be changed with this probability to one of the other founder values

missing.prob

Probability of missing data in final genotypes

full.prob

Probability of fully informative markers. Markers will be assigned with this probability to retain IBD genotypes from founders rather than being recoded into binary values. See details below for more information

keep.qtlgeno

Flag for whether to retain the QTL genotypes as a component in the output mpcross object

transpos

Positions of potential translocation (vector)

transval

Which founder carries the translocation

map.function

Map function for conversion of linkage map into recombination fractions. Default is "haldane"

seed

Random seed for generation of data

Input founder genotypes (optional) - otherwise generated randomly

founderld

Flag for whether to generate founder genotypes in linkage equilibrium (FALSE=default) or according to recombination map (TRUE)

Value

ibd: Fully informative founder genotypes for all markers
qtlgeno: If argument keep.qtlgeno is TRUE then QTL genotypes will be retained

Details

Data are initially generated by transmitting founder genotypes down through the pedigree to the finals. Errors, missing data, and binary alleles are then overlaid on this data (stored in $ibd). If founderld==FALSE, binary alleles are generated at each locus with probability 0.25 that one founder will have the allele; 0.50 that two founders will have the allele; and 0.25 that three founders will have the allele. The founders with the allele are randomly selected after the number of founders with the allele has been simulated. If founderld==TRUE then some markers may be monomorphic and will need to be removed from the resulting object using clean.mpcross.

Note that if founder genotypes are input they should be coded as follows: DArT markers take values in 0,1 SNP markers take values in 0,2 All other markers take some other set of values.

Examples

Run this code

map <- sim.map(len=100, n.mar=11, eq.spacing=TRUE, include.x=FALSE)
sim.ped <- sim.mpped(4, 1, 500, 6, 1)
sim.dat <- sim.mpcross(map=map, pedigree=sim.ped, qtl=matrix(data=c(1, 50, .4, 0, 0, 0), nrow=1, ncol=6, byrow=TRUE), seed=1)