SnpMatrix-class: Class "SnpMatrix"

In NikNakk/snpStats: SnpMatrix and XSnpMatrix classes and methods

Description

This class defines objects holding large arrays of single nucleotide polymorphism (SNP) genotypes generated using array technologies.

Objects from the Class

Objects can be created by calls of the form new("SnpMatrix", x) where x is a matrix with storage mode "raw". Chips (usually corresponding to samples or subjects) define rows of the matrix while polymorphisms (loci) define columns. Rows and columns will usually have names which can be used to link the data to further data concerning samples and SNPs

Slots

.Data:

Object of class "matrix" and storage mode raw Internally, missing data are coded 0 and SNP genotypes are coded 1, 2 or 3. Imputed values may not be known exactly. Such uncertain calls are grouped by probability and represented by codes 4 to 253

Extends

Class "matrix", from data part. Class "structure", by class "matrix". Class "array", by class "matrix". Class "vector", by class "matrix", with explicit coerce. Class "vector", by class "matrix", with explicit coerce.

Methods

[]

signature(x = "SnpMatrix", i = "ANY", j = "ANY", drop = "missing"): subset operations

cbind2

signature(x = "SnpMatrix", y = "SnpMatrix"): S4 generic function to provide cbind() for two or more matrices together by column. Row names must match and column names must not coincide. If the matrices are of the derived class XSnpMatrix-class, the diploid slot values must also agree

coerce

signature(from = "SnpMatrix", to = "numeric"): map to numeric values 0, 1, 2 or, for uncertain assignments, to the posterior expectation of the 0, 1, 2 code

coerce

signature(from = "SnpMatrix", to = "character"): map to codes "A/A", "A/B", "B/B", ""

coerce

signature(from = "matrix", to = "SnpMatrix"): maps numeric matrix (coded 0, 1, 2 or NA) to a SnpMatrix

coerce

signature(from = "SnpMatrix", to = "XSnpMatrix"): maps a SnpMatrix to an XSnpMatrix. Ploidy is inferred from the genotype data since haploid genotypes should always be coded as homozygous. After inferring ploidy, heterozygous calls for haploid genotypes are set to NA

is.na

signature(x = "SnpMatrix"): returns a logical matrix indicating whether each element is NA

rbind2

signature(x = "SnpMatrix", y = "snp.matrix"): S4 generic function to provide rbind() for two or more matrices by row. Column names must match and duplicated row names prompt warnings

show

signature(object = "SnpMatrix"): shows the size of the matrix (since most objects will be too large to show in full)

summary

signature(object = "SnpMatrix"): returns summaries of the data frames returned by row.summary and col.summary

is.na

signature(x = "SnpMatrix"): returns a logical matrix of missing call indicators

switch.alleles

signature(x = "SnpMatrix", snps ="ANY"): Recode specified columns of of the matrix to reflect allele switches

Note

This class requires at least version 2.3 of R

Author(s)

David Clayton [e-mail]

See Also

XSnpMatrix-class

Examples

data(testdata)
summary(Autosomes)

# Just making it up - 3-10 will be made into NA during conversion
snps.class<-new("SnpMatrix", matrix(1:10))
snps.class
if(!isS4(snps.class)) stop("constructor is not working")

pretend.X <- as(Autosomes, 'XSnpMatrix')
if(!isS4(pretend.X)) stop("coersion to derived class is not S4")
if(class(pretend.X) != 'XSnpMatrix') stop("coersion to derived class is not working")

pretend.A <- as(Xchromosome, 'SnpMatrix')
if(!isS4(pretend.A)) stop("coersion to base class is not S4")
if(class(pretend.A) != 'SnpMatrix') stop("coersion to base class is not working")

# display the first 10 snps of the first 10 samples
print(as(Autosomes[1:10,1:10], 'character'))

# convert the empty strings (no-calls) explicitly to "NC" before
# writing to an (anonymous and temporary) csv file
csvfile <- tempfile()
write.csv(file=csvfile, gsub ('^$', 'NC',
                              as(Autosomes[1:10,1:10], 'character')
                             ), quote=FALSE)
unlink(csvfile)

NikNakk/snpStats documentation built on May 7, 2019, 6:18 p.m.