dosagesJagsMix function

Compute dosages under specified Bayesian mixture model

Computes and returns estimated dosages under specified model using posterior probabilities derived from mcmc chains by the proportion of samples in each dosage class.

dosagesJagsMix(mcmc.mixture, jags.control, seg.ratio, chain = 1,
max.post.prob = TRUE, thresholds = c(0.5, 0.6, 0.7, 0.8, 0.9, 0.95,
0.99), print = FALSE, print.warning = TRUE, index.sample = 20)

Arguments

• mcmc.mixture: Object of type segratioMCMC

  produced by coda usually by using readJags

• jags.control: Object of class jagsControl for setting up JAGS command file

• seg.ratio: Object of class segRatio contains the segregation ratios for dominant markers and other information such as the number of dominant markers per individual

• chain: Which chain to use when compute dosages (Default: 1)

• max.post.prob: Logical for producing dose allocations based on the maximum posterior probability (Default: TRUE)

• thresholds: Numeric vector of thresholds for allocating dosages when the posterior probabilty to a particular dosage class is above the threshold

• print: Logical indicating whether or not to print intermediate results (Default: FALSE)

• print.warning: Logical to print warnings if there is more than one marker with the maximum posterior probability

• index.sample: Numeric vector indicating which markers to print if print is TRUE. If index.sample is of length 1 then a random sample of size index.sample is selected

Returns

An object of class dosagesMCMC is returned with components: - p.dosage: Matrix of posterior probabilities of dosages for each marker dosage

• dosage: Matrix of allocated dosages based on posterior probabilities. The columns correspond to different 'thresholds' and if requested, the last column is allocated on basis of max.post

• thresholds: vector of cutoff probabilities for dosage class

• chain: Chain used to compute dosages

• max.post: maximum dosage posterior probabilties for each marker

• index.sample: Numeric vector indicating which markers to print if print is TRUE. If index.sample is of length 1 then a random sample of size index.sample is selected

Author(s)

Peter Baker p.baker1@uq.edu.au

See Also

dosagesMCMC readJags

Examples

## simulate small autooctaploid data set
a1 <- sim.autoMarkers(8,c(0.7,0.2,0.1),n.markers=100,n.individuals=50)

## compute segregation ratios
sr <-  segregationRatios(a1$markers)

## set up model, priors, inits etc and write files for JAGS
x <- setModel(3,8)
x2 <- setPriors(x)
dumpData(sr, x)
inits <- setInits(x,x2)
dumpInits(inits)
writeJagsFile(x, x2, stem="test")

## Not run:

## run JAGS
small <- setControl(x, burn.in=200, sample=500)
writeControlFile(small)
rj <- runJags(small)  ## just run it
print(rj)

## read mcmc chains and produce dosage allocations
xj <- readJags(rj)
dd <- dosagesJagsMix(xj, small, sr)
print(dd)
## End(Not run)