bicreg.models {BayesQTLBIC}R Documentation

BIC analysis for a fixed set of models

Description

bicreg.models evaluates posterior probabilities based on the BIC criterion and prior probabilities for a fixed set of models, typically obtained by sampling from sets of models from separate analyses of individual chromosomes.

Usage

bicreg.models(x,y,wt = rep(1, length(y)),which,intercept=TRUE,add.null.model=TRUE,
                          n=length(y)/num.imputations,num.imputations=1,delta=1,
                          p.sg=1,prior=0.5,eval.markers=TRUE,neval=NULL)

Arguments

x Matrix of independent variables, based on marker genotypes, often from a single chromosome.
y Vector of values for the dependent variable (trait values).
wt Vector of weights for regression.
which Matrix of logical values corresponding to a set of models, the (i,j) element is TRUE if and only if the jth variable is selected in the ith model.
intercept Add an intercept term.
add.null.model Add the NULL model.
n Original sample size, before multiple imputations.
num.imputations Number of imputations used to construct x, y.
prior Vector or scalar specifying prior probabilities per marker for a QTL to be in the vicinity of the marker; generally proportional to the distance to flanking markers and total number of QTL expected genome. Defaults to 0.5 which is usually too high.
delta Adjustment factor for the penalty term in the BIC criterion, default is no adjustment delta=1; (Cf. Broman and Speed 2002); not needed if using subjective prior probabilities and sample size is ample (p.sg=1 and n >= 100; Ball 2007).
p.sg Proportion p.sg/2 of each tail is genotyped if selective genotyping is being used; default 1, corresponding to fully genotyped population.
eval.markers Evaluate model averaged estimates for marker effects (effects of allelic substitution).
neval Use neval top models on which to evaluate model averaged estimates of marker effects, default NULL, use all models.

Details

Provides posterior probabilities for a fixed set of linear models representing alternative QTL genetic architectures. Provides Bayesian model averaged estimates for effects of QTL or effects of allelic substitution for markers which may be linked to QTL.

Value

bicreg.models returns an object of class bicreg.qtl.

Author(s)

R.D. Ball, (rod.ball@AT@scionresearch.com)

References

Ball, R. D. 2001: Bayesian methods for QTL mapping based on model selection: approximate analysis using the Bayesian Information Criterion. Genetics 159: 1351–1364.

See Also

bicreg.qtl,sample.bicreg.qtl.models

Examples

## Not run: 
data(ex3n300a.data)
chrom <- rep(1:12,rep(16,12))
marker <- rep(1:16,12)
x <- sapply(ex3n300a.data$Markers,c)
y <- ex3n300a.data$Trait$t1
nchrom <- length(sort(chrom.levels <- unique(chrom)))

quick.demo <- TRUE
if(quick.demo){
  nc <- 2; nsim <- 20;x <- x[,chrom %in% 1:2];
  chrom <- chrom[chrom %in% 1:2]
}else{nc <- 12; nsim <- 200}
chrom.fits <- list()
for(ii in seq(along=chrom.levels[1:nc])){
  cat(paste("*** chromosome",ii,"***","\n"))
  ci <- chrom.levels[ii]
  chrom.sel <- chrom==ci
  chrom.fits[[ii]] <- bicreg.qtl(x[,chrom.sel],y, prior=0.1,nbest=20,nvmax=3)
}
mWhich <- sample.bicreg.qtl.models(chrom.fits,nsim=nsim)
mres <- bicreg.models(x=x,y=y,which=mWhich,prior=0.1)
summary(mres,nbest=38,min.marker.prob=0.05)
## End(Not run)

[Package BayesQTLBIC version 1.0-0 Index]