qtlnet (version 1.5.4)

cyclica: Cyclic graph (a) example

Description

We use a Gibbs sampling scheme to generate a data-set with 200 individuals (according with cyclic graph (a)). Each phenotype is affected by 3 QTLs. We fixed the regression coefficients at 0.5, error variances at 0.025 and the QTL effects at 0.2, 0.3 and 0.4 for the three F2 genotypes. We used a burn-in of 2000 for the Gibbs sampler.

Usage

data(cyclica)

Details

For cyclic graphs, the output of the qdg function computes the log-likelihood up to the normalization constant (un-normalized log-likelihood). We can use the un-normalized log-likelihood to compare cyclic graphs with reversed directions (they have the same normalization constant). However we cannot compare cyclic and acyclic graphs.

References

Chaibub Neto et al. (2008) Inferring causal phenotype networks from segregating populations. Genetics 179: 1089-1100.

See Also

sim.cross, sim.geno, sim.map, skeleton, qdg, graph.qdg, generate.qtl.pheno

Examples

# NOT RUN {
bp <- matrix(0, 6, 6)
bp[2,1] <- bp[4,2] <- bp[4,3] <- bp[5,4] <- bp[2,5] <- bp[6,5] <- 0.5
stdev <- rep(0.025, 6)

## Use R/qtl routines to simulate.
set.seed(3456789)
mymap <- sim.map(len = rep(100,20), n.mar = 10, eq.spacing = FALSE,
  include.x = FALSE)
mycross <- sim.cross(map = mymap, n.ind = 200, type = "f2")
mycross <- sim.geno(mycross, n.draws = 1)

cyclica.qtl <- generate.qtl.markers(cross = mycross, n.phe = 6)
mygeno <- pull.geno(mycross)[, unlist(cyclica.qtl$markers)]

cyclica.data <- generate.qtl.pheno("cyclica", cross = mycross, burnin = 2000,
  bq = c(0.2,0.3,0.4), bp = bp, stdev = stdev, geno = mygeno)
save(cyclica.qtl, cyclica.data, file = "cyclica.RData", compress = TRUE)

data(cyclica)
out <- qdg(cross=cyclica.data, 
		phenotype.names=paste("y",1:6,sep=""),
		marker.names=cyclica.qtl$markers, 
		QTL=cyclica.qtl$allqtl, 
		alpha=0.005, 
		n.qdg.random.starts=10,
		skel.method="pcskel")


gr <- graph.qdg(out)
gr
plot(gr)
# }