recodeSNPs {scrime}

Recoding of SNP Values

Description

Recodes the values used to specify the genotypes of the SNPs to other values. Such a recoding might be required to use other functions contained in this package.

Usage

recodeSNPs(mat, first.ref = FALSE, geno = 1:3, snp.in.col = FALSE)

Arguments

mat

a matrix or data frame consisting of character strings of length 2 that specify the genotypes of the SNPs. Each of these character strings must be a combination of the letters A, T, C, and G. Missing values can be specified by "NN" or NA. Depending on snp.in.col it is assumed that each row of mat represents a SNP and each column a variable (snp.in.col = FALSE), or vice versa.

first.ref

does the first letter in the string coding the heterozygous genotype always stands for the more frequent allele? E.g., codes "CC" for the homozygous reference genotype if the genotypes of a SNP are coded by "CC", "CG" and "GG"? If TRUE, the value made up only of this first letter is set to geno[1], and the value made up only of the second letter is set to geno[3]. If FALSE, it is evaluated rowwise which of the homozygous genotypes has the higher frequency and the more often occuring value is set to geno[1], and the other to geno[3].

geno

a numeric or character vector of length 3 giving the three values that should be used to recode the genotypes. By default, geno = 1:3 which is the coding, e.g., required by rowChisqStats or pamCat.

snp.in.col

does each column of mat correspond to a SNP (and each row to an array)? If FALSE, it is assumed that each row represents a SNP, and each column an array.

Value

A matrix of the same size as mat containing the recoded genotypes. (Missing values are coded by NA).

See Also

recodeAffySNP, snp2bin

Examples

## Not run: 
# Generate an example data set consisting of 5 rows and 12 columns,
# where it is assumed that each row corresponds to a SNP.

mat <- matrix("", 10, 12)
mat[c(1, 4, 6),] <- sample(c("AA", "AT", "TT"), 18, TRUE)
mat[c(2, 3, 10),] <- sample(c("CC", "CG", "GG"), 18, TRUE)
mat[c(5, 8),] <- sample(c("GG", "GT", "TT"), 12, TRUE)
mat[c(7, 9),] <- sample(c("AA", "AC", "CC"), 12, TRUE)
mat

# Recode the SNPs

recodeSNPs(mat)

# Recode the SNPs by assuming that the first letter in
# the heterogyzous genotype refers to the major allele.

recodeSNPs(mat, first.ref = TRUE)


## End(Not run)