sommer (version 4.1.6)

atcg1234: Letter to number converter

Description

This function was designed to help users to transform their data in letter format to numeric format. Details in the format are not complex, just a dataframe with markers in columns and individuals in rows. Only markers, NO extra columns of plant names etc (names of plants can be stored as rownames). The function expects a matrix of only polymorphic markers, please make sure you clean your data before using this function. The apply function can help you identify and separate monomorphic from polymorphic markers.

Usage

atcg1234(data, ploidy=2, format="ATCG", maf=0, multi=TRUE, 
        silent=FALSE, by.allele=FALSE, imp=TRUE, ref.alleles=NULL)

Arguments

data

a dataframe with markers in columns and individuals in rows. Preferable the rownames are the ID of the plants so you don't lose track of what is what.

ploidy

a numeric value indicating the ploidy level of the specie. The default is 2 which means diploid.

format

one of the two possible values allowed by the program "ATCG", which means your calls are in base-pair-letter code, i.e. "AT" in a diploid call, "AATT" tetraploid etc (just example). Therefore possible codes can be "A", "T", "C", "G", "-" (deletion), "+" (insertion). Alternatively "AB" format can be used as well. Commonly this depends from the genotyping technologies used, such as GBS or microarrays. In addition, we have enabled also the use of single-letter code used by Cornell, i.e. A=AA, C=CC, T=TT, G=GG, R=AG, Y=CT, S=CG, W=AT, K=GT, M=AC. In the case of GBS code please make sure that you set the N codes to regular NAs handled by R. The "ATCG" format also works for the bi-allelic marker codes from join map such as "lm", "ll","nn", "np","hh","hk","kk"

maf

minor allele frequency used to filter the SNP markers, the default is zero which means all markers are returned in numeric format.

multi

a TRUE/FALSE statement indicating if the function should get rid of the markers with more than 2 alleles. If FALSE, which indicates that if markers with multiple alleles are found, the alternate and reference alleles will be the first 2 alleles found. This could be risky since some alleles will be masked, i.e. AA AG AT would take only A and G as reference and alternate alleles, converting to numeric format 2 1 1, giving the same effect to AG and AT which could be a wrong assumption. The default is TRUE, removes markers with more than two alleles.

silent

a TRUE/FALSE value indicating if a progress bar should be drawn for each step of the conversion. The default is silent=FALSE, which means that we want progress bar to be drawn.

by.allele

a TRUE/FALSE value indicating if the program should transform the data in a zero/one matrix of presence/absense per allele. For example, a marker with 3 alleles A,T,C in a diploid organism will yield 6 possible configurations; AA, AT, AC, TT, TC, CC. Therefore, the program would create 3 columns for this marker indicating the presence/absence of each allele for each genotype.

imp

a TRUE/FALSE value indicating if the function should impute the missing data using the median for each marker. If FALSE, then the program will not impute.

ref.alleles

a matrix with reference alleles to be used for the conversion. The matrix should have as many columns as markers with reference alleles and with 2 rows, being the first row the alternate allele (Alt) and the second row the reference allele (Ref). Rownames should be "Alt" and "Ref" respectively. If not provided the program will decide the reference allele.

Value

$data

a new dataframe of markers in numeric format with markers in columns and individuals in rows.

See Also

The core function of the package mmer

Examples

# NOT RUN {
data(DT_polyploid)
genotypes <- GT_polyploid
genotypes[1:5,1:5] # look the original format

####=================================================####
#### convert markers to numeric format polyploid potatoes
####=================================================####
# numo <- atcg1234(data=genotypes, ploidy=4)
# numo$M[1:5,1:5]

####=================================================####
#### convert markers to numeric format diploid rice lines
#### single letter code for inbred lines from GBS pipeline
#### A=AA, T=TT, C=CC, G=GG 
####=================================================####
# data(DT_rice)
# X <- GT_rice; X[1:5,1:5]; dim(X)
# numo2 <- atcg1234(data=X, ploidy=2)
# numo2$M[1:5,1:5]
# }