New function findExclusions() for identifying incompatible markers in identification cases.
powerPlot() gains a logical argument jitter, which can be switched on to avoid overplotting.
checkPairwise() gains an argument excludeInbred, which is TRUE by default. This is sensible since the plot shows estimated kappa coefficients, which are well-behaved only for pairs of noninbred individuals.
forrel now requires R version 4.1 and recent versions of pedtools and ribd. This allowed many simplifications in code and examples.
Added scales as a suggested package.
The new function ibdEstimate() replaces the previous IBDestimate() (note the name change). This is a complete rewrite, which optimises the log-likelihood using a projected gradient descent algorithm, combined with a version of Armijo line search.
The function ibdBootstrap() replaces the previous kappaBootstrap() and deltaBootstrap(), and is considerably faster. This function implements both parametric and non-parametric bootstrap, controlled with the method parameter.
The output of ibdEstimate() now has a class attribute “ibdEst”, with its own print and subsetting methods.
kinshipLR() now handles linked markers by wrapping MERLIN.
New functions kappaBootstrap() and deltaBootstrap() for assessing the uncertainty of pairwise relatedness estimates.
New function randomPersonEP() handling a common special case of exclusionPower().
forrel now depends on version 0.9.6 (or later) of pedtools.
Deprecated arguments id.labels and frametitles in missingPersonPlot() has been removed.
Implement parallelisation in profileSim().
Partial rewrite of kinshipLR(), including new argument source.
Added the NorwegianFrequencies dataset, containing allele frequencies for 35 STR markers.
New function missingPersonLR().
New function checkPairwise() replaces the (long obsolete) examineKinships().
New functions markerSimParametric() and profileSimParametric() for simulating marker data for two individuals with given kappa (or condensed identity) coefficients.
profileSim(), fix bug resulting in identical seeds given to each parallel cluster.readFam() now has a parameter Xchrom which can be used to indicate that the markers included in the file are on the X chromosome
MPPsims() is more flexible, and allows subsetting of its output.
powerPlot() is more flexible and allows finer control of the plot contents
readFam(). It is more robust now, and fails gracefully in certain situations which cannot currently be handled (e.g. if the file contains twins).