merge-methods {polysat}

Merge Two Genotype Objects into One

Description

The generic function merge has methods defined in polysat to merge two genotype objects of the same class. Each method has optional samples and loci arguments for specifying subsets of samples and loci to be included in the merged object. Each method also has an optional overwrite argument to specify which of the two objects should not be used in the case of conflicting data.

Usage

merge(x, y, ...)

Arguments

x

One of the objects to be merged. For the methods defined for polysat this should be of class "gendata" or one of its subclasses.

y

The other object to be merged. Should be of the same class as x. y@Ploidies must also be of the same class as x@Ploidies.

...

Additional arguments specific to the method.

Methods

The methods for merge in polysat have four additional arguments: objectm, samples, loci, overwrite.

The samples and loci arguments can specify, using character vectors, a subset of the samples and loci found in x and y to write to the object that is returned.

If overwrite = "x", data from the second object will be used wherever there is contradicting data. Likewise if overwrite = "y", data from the first object will be used wherever there is contradicting data. If no overwrite argument is given, then any contradicting data between the two objects will produce an error indicating where the contradicting data were found.

The objectm argument is primarily for internal use (most users will not need it). If this argument is not provided, a new genotype object is created and data from x and y are written to it. If objectm is provided, this is the object to which data will be written, and the object that will be returned.

signature(x = "genambig", y = "genambig")

This method merges the genotype data from x and y. If the missing data symbols differ between the objects, overwrite is used to determine which missing data symbol to use, and all missing data symbols in the overwritten object are converted. If overwrite is not provided and the missing data symbols differ between the objects, an error will be given. The genotypes are then filled in. If certain sample*locus combinations do not exist in either object (x and y have different samples as well as different loci), missing data symbols are left in these positions. Again, for genotypes, overwrite determines which object to preferentially use for data and whether to give an error if there is a disagreement.

The merge method for gendata is then called.

signature(x = "genbinary", y = "genbinary")

This method also merges genotype data for x and y, then calls the method for gendata. Missing, Present, and Absent are checked for consistency between objects similarly to what happens with Missing in the genambig method. Genotypes are then written to the merged object, and consistency between genotypes is checked.

signature(x = "gendata", y = "gendata")

This method merges data about ploidy, repeat length, and population identity, as well as writing one or both dataset descriptions to the merged object.

The same population numbers can have different meanings in PopInfo(x) and PopInfo(y). The unique PopNames are used instead to determine population identity, and the PopInfo numbers are changed if necessary. Therefore, it is important for identical populations to be named the same way in both objects, but not important for identical populations to have the same number in both objects.