geno: Create a geno Object.

Description

Creates an object of class geno containing a matrix of one-column-per-marker formated genotype data, the type of kernel, and the weights to be used to create the kernel matrix. This function is used to simplify the input structure of the call to KITdesign.

Usage

geno(mat, kernel = "linear", weights = NULL, inheritMode = NA)

Arguments

mat

Design matrix for genotype data in either one- or two-column-per-marker format. Object can be a vector, matrix, or data.frame. Missing values can be coded as either 9 or NA.

kernel

Character object indicating type of kernel. Must be one of 'ibs', 'linear', 'quadratic'. For linear kernel, the constant is taken to be zero; for quadratic kernel, it is one.

weights

Weights, if any, to be used in generating kernel. Object can be a vector or matrix. If vector and kernel is not ibs, weights are converted to a diagonal matrix.

inheritMode

For two-column-per-marker format, the inheritance mode to be used to convert to one-column-per-marker format. Must be one of 'add', 'dom', 'rec' indicating additive, dominant, or recessive, respectively.

Value

Returns an object of class geno containing the one-column-per-marker genotype data, the weights to be used in generating the kernel, and the type of kernel to be generated.

Details

There are two conventions for genotype data: one- and two-column-per-marker formats. Either format can be used. If genotype data is in one-column format, inheritMode = NA. If genotype data is in two-column format, inheritMode = add/dom/rec.

Non-integer imputed values can be provided. If it is determined that non-integer values are provided and that the kernel was specified as `ibs,' the kernel will be reset to `linear.' In addition, if provided in the two-column format, inheritMode will be set to `add.'

Examples

Run this code

# NOT RUN {
  gdata <- matrix(sample(0:1,40,replace=TRUE,prob=c(0.9,0.1)),ncol=4L)

  geno(mat = gdata, kernel = "ibs", 
       weights = NULL, inheritMode="add")

# }