cross2int {wgaim}

Convert a cross genetic object to an interval object

Description

Converts an object of class "cross" to an object with class "interval". The function also imputes missing markers.

Usage

cross2int(object, impute = "MartinezCurnow", consensus.mark = TRUE,
     id = "id", subset = NULL)

Arguments

object

an object of class "cross" that inherits one of the class structures "bc", "dh", "f2", "riself".

impute

a character string determining how missing values in the linkage map should be imputed. If "Broman", then missing values are imputed according to Bromans rules. If "MartinezCurnow" then missing values are imputed according to the rules of Martinez & Curnow (1994) (see reference list). The default is "MartinezCurnow" (see Details).

consensus.mark

logical value. If TRUE co-locating marker sets are condensed to form consensus markers (see Details). Defaults to TRUE.

id

a character string or name of the unique identifier for each row of genotype data (see Details). Defaults to "id"

subset

a possible character vector naming the subset of chromosomes to be returned. Defaults to NULL implying return all chromosomes.

Details

This function provides the conversion of genetic data objects that have already been generated using read.cross() from Bromans qtl package, to "interval" objects ready for use with wgaim. Users should be aware that this function is restricted to certain populations. object must inherit one of the class structures "bc", "dh", "f2", "riself".

During the conversion process three important linkage map attributes are assessed.

1. The map may be subsetted using the subset argument

2. If consensus.mark = TRUE then co-located marker sets are reduced to form single consensus markers before missing values are imputed. The marker similarity is determined by the genetic distances that are given in the map component for each linkage group. If a set of markers co-locate the name of the first marker is chosen and a single consensus marker is determined by coalescing the genetic information from all markers in the set. A "(C)" is placed after the marker name for easy identification. The markers removed from each set are returned with the object and placed under "colocated.markers" for inspection if required.

3. Missing values are imputed according to the argument given by impute. This imputation results in a complete version of the marker data for each chromosome which is then used to create the interval data component "interval.data". The complete marker data for each chromosome can be obtained from the "imputed.data" element of the returned list. It is therefore also possible to perform whole genome marker analysis using wgaim. See wgaim.asreml for more details.

Value

a list of class "cross" that also inherits the class "interval". The list contains the following components

geno

A list with elements named by the corresponding names of the chromosomes. Each chromosome is itself a list with six elements: "data" is the actual estimated map matrix with rows as individuals named by "id" and markers as columns; "map" is a vector of marker positions on the corresponding chromosome; "imputed.data" is identical to "data" matrix but with all NAs replaced by imputed values according to the rules of "impute"; "dist" contains the genetic distance between adjacent markers or the genetic distances of the intervals; "theta" contains the recombination fractions for each interval; "interval.data" contains the recalculated intervals based on the recombination fractions and the missing marker information.

colocated.markers

If consensus.mark = TRUE, a four column data frame containing stacked binned sets of co-located markers. In each binned set the first marker is the unique consensus marker name used in the linkage map and the others are the co-located marker names that were omitted. Additionally for each binned set, the data frame also contains linkage group identification and marker position information.

pheno

A data.frame of phenotypic information with rows as individuals read in from read.cross. A copy of the column named by the "id" argument can be found here (see the help for the read.cross() function).

Author(s)

Julian Taylor and Ari Verblya

References

Martinez, O., Curnow. R. N. (1994) Missing markers when estimating quantitative trait loci using regression mapping. Heredity, 73, 198-206.

Julian Taylor, Arunas Vebyla (2011). R Package wgaim: QTL Analysis in Bi-Parental Populations Using Linear Mixed Models. Journal of Statistical Software, 40(7), 1-18. URL http://www.jstatsoft.org/v40/i07/.

Verbyla, A. P., Cullis, B. R., Thompson, R (2007) The analysis of QTL by simultaneous use of the full linkage map. Theoretical And Applied Genetics, 116, 95-111.

See Also

read.cross

Examples

## Not run: 
# read in linkage map from a rotated .CSV file with "id" as the
# identifier for each unique row

wgpath <- system.file("extdata", package = "wgaim")
genoSxT <- read.cross("csvr", file="genoSxT.csv", genotypes=c("AA","BB"),
         na.strings = c("-", "NA"), dir = wgpath)
genoSxT <- cross2int(genoSxT, impute="MartinezCurnow", id = "id")

# plot linkage map

linkMap(genoSxT, cex = 0.5)


## End(Not run)