reordXlist: Reorders Design Matrices

In MasterBayes: ML and MCMC Methods for Pedigree Reconstruction and Analysis

Reorders Design Matrices

Description

Reorders design matrices so excluded parents appear last, and high probability parents appear first, thus increasing computational efficiency.

Usage

  reordXlist(X.list, marker.type="MSW")

Arguments

X.list	list of design matrices for each offspring derived using getXlist. Mismatch information must be present (see mismatches)
marker.type	"MSW" or "MSC" for co-dominant markers with Wang's (2004) model of genotyping error or CERVUS's model of genotyping error (Kalinowski, 2006; Marshall, 1998) or "AFLP" for dominant markers (Hadfield, 2009).

Details

The design matrices are reordered by the number of mismatches between a parent and offspring for codominant markers, and by the probability of the offspring genotype conditional on parent genotype for dominant markers.

Value

X.list for which parents are reordered

Note

If a GdataPed object is passed to getXlist then the design matrices will be reordered by default.

Author(s)

Jarrod Hadfield [e-mail]

See Also

MCMCped

Examples

## Not run: 
data(WarblerG)
A<-extractA(WarblerG)

ped<-matrix(NA, 5,3)
ped[,1]<-1:5
ped[,2]<-c(rep(NA, 4), 3)
ped[,3]<-c(rep(NA, 4), 4)

genotypes<-simgenotypes(A, ped=ped)

sex<-c("Female", "Male", "Female", "Male","Female")
offspring<-c(0,0,0,0,1)

data<-data.frame(id=ped[,1], sex, offspring)

var1<-expression(varPed(x="offspring", restrict=0))
PdP<-PdataPed(formula=list(var1), data=data)

X.list<-getXlist(PdP)
# creates design matrices for offspring (in this case indivdiual "5")

X.list<-mismatches(X.list, G=genotypes$Gobs)
X.list<-fillX.G(X.list, A=A, G=genotypes$Gobs)

X.list.reord<-reordXlist(X.list)

# The design matrices for the genetic likelihoods are reordered
# by the number of mismatches.  The true parental combination
# now appears first rather than last.

X.list$X$"5"$G
X.list.reord$X$"5"$G

## End(Not run)

MasterBayes documentation built on June 22, 2022, 5:06 p.m.