get.likely.genotypes.peaks: likeLTD::get.likely.genotypes.peaks
In likeLTD: Tools to Evaluate DNA Profile Evidence

Description Usage Arguments Details Value See Also Examples

Creates a list of the most likely genotypes at each locus, and the most likely whole-profile genotype for peak height data.

1 2	get.likely.genotypes.peaks(hypothesis,params,results, posterior=FALSE,joint=FALSE,prob=ifelse(joint==FALSE,0.1,0.05))

`hypothesis`	Hypothesis object created by either `defence.hypothesis.peaks` or `prosecution.hypothesis.peaks` that was used for optimisation with `evaluate.peaks`.
`params`	Parameters object created by `optimisation.params.peaks` that was used for optimisation with `evaluate`.
`results`	Either prosecution or defence results returned by `evaluate.peaks` e.g. `results$Pros` or `results$Def`.
`posterior`	Logical indicating whether to return all genotype probabilities, rather than just the most likely.
`joint`	Logical indicating whether or not to return joint genotypes and probabilities. If FALSE, marginal genotypes and probabilities are returned instead.
`prob`	Probability threshold for single-locus genotype probabilities. Defaults to 0.1 if returning marginal probabilities, and 0.05 if returning joint probabilities.

Either joint or marginal genotypes and genotype probabilities are given. Locus-specific genotypes are only given if their probabilty exceeds prob. The most likely whole-profile genotype is given, regardless of the probability threshold at each locus. Joint probabilities give the probability of a multi-contributor genotype, whereas marginal probabilities give the probability of a single contributor, summing over all the possible genotypes for all other contributors.

`locusSpecific`	Locus genotypes and probabilities which are greater than `prob` for each contributor of `joint=FALSE`, or for all contributors if `joint=TRUE`.
`topGenotype`	Most likely whole-profile genotype for all contributors if `joint=TRUE` or for each contributor separately if `joint=FALSE`. The probability of each genotype is also given.

defence.hypothesis.peaks, prosecution.hypothesis.peaks, optimisation.params.peaks,evaluate.peaks

## Not run: 
# datapath to example files
datapath = file.path(system.file("extdata", package="likeLTD"),"laboratory")

# File paths and case name for allele report
admin = pack.admin.input.peaks(
            peaksFile = file.path(datapath, 'laboratory-CSP.csv'),
            refFile = file.path(datapath, 'laboratory-reference.csv'),
            caseName = "Laboratory",
            detectionThresh = 20
             )

# Enter arguments
args = list(
        nUnknowns = 1
        )

# Create hypotheses
hypP = do.call(prosecution.hypothesis.peaks, append(admin,args))
hypD = do.call(defence.hypothesis.peaks, append(admin,args))

# Get parameters for optimisation
paramsP = optimisation.params.peaks(hypP)
paramsD = optimisation.params.peaks(hypD)

# reduce number of iterations for demonstration purposes
paramsP$control$itermax=25
paramsD$control$itermax=25

# Run optimisation
# n.steps and converge set for demonstration purposes
results = evaluate.peaks(paramsP, paramsD, n.steps=1, 
      converge=FALSE)

# get most likely marginal genotypes under defence
get.likely.genotypes.peaks(hypD,paramsD,
      results$Def)

# get most likely joint genotypes under defence
gensJoint = get.likely.genotypes.peaks(hypD,paramsD,
      results$Def,joint=TRUE)

# get posterior likelihoods for all genotype combinations
  gensPosterior = get.likely.genotypes.peaks(hypD,paramsD,
      results$Def,posterior=TRUE)

## End(Not run)