hypredNewQTL: Assign new QTLs to a hypredGenome object
In timflutre/hypred: Simulation of Genomic Data in Applied Genetics

Description Usage Arguments Details Value References See Also Examples

A generic function to assign new QTLs to a "hypredGenome" object

hypredNewQTL(object, ...)
## S4 method for signature 'hypredGenome'
hypredNewQTL(object, new.id.add = NULL,
  new.id.dom = NULL, new.id.per.mar = NULL, new.eff.add = NULL,
  new.eff.dom = NULL)

`object`	an object of a class that holds genomic information on map positions of loci, typically an `"hypredGenome"` object.
`new.id.add`	integer vector giving the IDs of the QTL with additive effects (hence all QTL).
`new.id.dom`	integer vector giving the IDs of the QTL defined with `new.id.add` that have also a dominance effect.
`new.id.per.mar`	integer vector giving the IDs of the QTL defined with `new.id.add` that are taken to be perfect markers.
`new.eff.add`	numeric vector giving the additive effects of the QTL defined with `new.id.add`.
`new.eff.dom`	numeric vector giving the dominance effects of the QTL defined with `new.id.dom`.
`...`	Methods may require further arguments.

The number of loci, as defined through the function hypredGenome always remains constant. What can be changed is the number of these that are QTL.

The number of QTL, as well as QTL with dominance effect, must be the same for all chromosomes. Also the number of perfect markers must be the same for all chromosomes. There are some validity checks included to assure that. However, simply setting effects of some of the QTL to 0, is an indirect way to have different numbers of QTL per chromosome.

Only QTL with an additive effect may have dominance effects assigned. Hence, new.id.dom must be a subset of new.id.add. However, setting the additive effect of a QTL with dominance to zero is again an indirect way to have QTL with pure dominance. Of course, only QTL can be perfect markers, hence new.id.per.mar must be a subset of new.id.add as well.

The vector given to new.id.add holds the IDs of the locis that are assigned QTLs. The ID of a loci is its index in the vector of all loci in the object. If, for example, the object hold 20 loci, then the ID of the loci with lowest map postion of chromosome 1 is 1, the next has ID 2 and so on. The ID of the loci with lowes map position of chromosome 2 is 11. So if one wants to assign QTLs to the third locus on chromosome 1 and the fifths on chromosome 2, the vector given to new.eff.add needs to be c(3, 15).

The same applies for the vectors given to new.id.dom and new.id.per.mar.

The effects are distributed to the QTL in the order in which they appear in the vectors given to new.eff.add and new.eff.dom. So, inline with above example, new.eff.add = c(0.1, -0.1) would assign an additive effect of 0.1 to the QTL on chromosome 1 and -0.1 to the QTL on chromosome 2.

If no dominance QTL or perfect markers are assigned, the corresponding argument needs to be NULL.

The additive effects of the QTL are interpreted as the homozygous effects (a) of the Falconer scale (i.e. half the difference between the two homozygous genotypes). The dominance effects are interpreted as the d of the Falconer scale (i.e. the difference between the heterozygous genotype from the mean of the two homozygous genotypes).

The interpretation and treatment of QTL is similar to SNPs. It is assumed that the QTL alleles are identical to each other, apart from one causative SNP mutation. This mutation allows to identify them, so that a 1 codes for the mutated allele and a 0 for the unmutated allele. These causative SNP are treated as unobserved by default, hence they are excluded from the design matrices created by function hypredCode. However, with the argument new.id.per.mar causative SNP can be turned into observed, perfect markers that are included in the design matrices.

A "hypredGenome" object, modified accordingly.

D. S. Falconer (1967) Introduction to Quantitative Genetics. Oliver and Boyd., London, 1967.

The function hypredGenome which creates the original object and the function hypredTruePerformance which combines the QTL information into genotypic values of individuals.

## two chromosomes of length 1 M and 10 SNP (loci) per chromosome

no_QTL <- hypredGenome(2, c(1.0, 1.0), 10)

summary(no_QTL)

## assign a QTL to the third loci on chromosome 1 and the fifths on
## chromosome 2.

## The additive effect of the first QTL is 0.1, the additive effect of
## the second QTL is -0.1. The QTLs don't show dominance, both are
## perfect markers. 

with_QTL <- hypredNewQTL(no_QTL,
                         new.id.add = c(3, 15),
                         new.id.dom = NULL, ## default
                         new.id.per.mar = c(3, 15),
                         new.eff.add = c(0.1, -0.1),
                         new.eff.dom = NULL ## default
                        )

summary(with_QTL)