GdsGenotypeReader-class: Class GdsGenotypeReader

Description Extends Constructor Accessors Author(s) See Also Examples

Description

The GdsGenotypeReader class is an extension of the GdsReader class specific to reading genotype data stored in GDS files. GDS files with both snp x scan and scan x snp dimensions are supported.

Extends

GdsReader

Constructor

GdsGenotypeReader(filename, genotypeDim, allow.fork=FALSE, ...):

filename must be the path to a GDS file or a gds object. The GDS file must contain the following variables:

  • 'snp.id': a unique integer vector of snp ids

  • 'snp.chromosome': integer chromosome codes

  • 'snp.position': integer position values

  • 'sample.id': a unique integer vector of scan ids

  • 'genotype': a matrix of bytes with dimensions ('snp','sample'). The byte values must be the number of A alleles : 2=AA, 1=AB, 0=BB.

The optional variable "snp.allele" stores the A and B alleles in a character vector with format "A/B".

Default values for chromosome codes are 1-22=autosome, 23=X, 24=XY, 25=Y, 26=M. The defaults may be changed with the arguments autosomeCode, XchromCode, XYchromCode, YchromCode, and MchromCode.

The constructor automatically detects whether the GDS file is in snp x scan or scan x snp order using the dimensions of snp.id and sample.id. In the case of GDS files with equal SNP and scan dimensions, genotypeDim is a required input to the function and can take values "snp,scan" or "scan,snp".

allow.fork is a logical to enable multiple forks to access the gds file simultaneously.

The GdsGenotypeReader constructor creates and returns a GdsGenotypeReader instance pointing to this file.

Accessors

In the code snippets below, object is a GdsGenotypeReader object. See GdsReader for additional methods.

nsnp(object): The number of SNPs in the GDS file.

nscan(object): The number of scans in the GDS file.

getSnpID(object, index): A unique integer vector of snp IDs. The optional index is a logical or integer vector specifying elements to extract.

getChromosome(object, index, char=FALSE): A vector of chromosomes. The optional index is a logical or integer vector specifying elements to extract. If char=FALSE (default), returns an integer vector. If char=TRUE, returns a character vector with elements in (1:22,X,XY,Y,M,U). "U" stands for "Unknown" and is the value given to any chromosome code not falling in the other categories.

getPosition(object, index): An integer vector of base pair positions. The optional index is a logical or integer vector specifying elements to extract.

getAlleleA(object, index): A character vector of A alleles. The optional index is a logical or integer vector specifying elements to extract.

getAlleleB(object, index): A character vector of B alleles. The optional index is a logical or integer vector specifying elements to extract.

getScanID(object, index): A unique integer vector of scan IDs. The optional index is a logical or integer vector specifying elements to extract.

getGenotype(object, snp=c(1,-1), scan=c(1,-1), drop=TRUE, use.names=FALSE, transpose=FALSE, ...): Extracts genotype values (number of A alleles). snp and scan indicate which elements to return along the snp and scan dimensions. They must be integer vectors of the form (start, count), where start is the index of the first data element to read and count is the number of elements to read. A value of '-1' for count indicates that the entire dimension should be read. If drop=TRUE, the result is coerced to the lowest possible dimension. If use.names=TRUE, names of the resulting vector or matrix are set to the SNP and scan IDs. Missing values are represented as NA. Genotypes are returned in SNP x scan order if transpose=FALSE, otherwise they are returned in scan x SNP order.

getGenotypeSelection(object, snp=NULL, scan=NULL, snpID=NULL, scanID=NULL, drop=TRUE, use.names=TRUE, order=c("file", "selection"), transpose=FALSE, ...): Extracts genotype values (number of A alleles). snp and scan may be integer or logical vectors indicating which elements to return along the snp and scan dimensions. snpID and scanID allow section by values of snpID and scanID. Unlike getGenotype, the values requested need not be in contiguous blocks. If order=="file", genotypes are returned in the order they are stored in the file. If order="selection", the order of SNPs and scans will match the index selection provided in snp and scan respectively. Other arguments are identical to getGenotype.

getVariable(object, varname, index, drop=TRUE, ...): Extracts the contents of the variable varname. The optional index is a logical or integer vector (if varname is 1D) or list (if varname is 2D or higher) specifying elements to extract. If drop=TRUE, the result is coerced to the lowest possible dimension. Missing values are represented as NA. If the variable is not found, returns NULL.

XchromCode(object): Returns the integer code for the X chromosome.

XYchromCode(object): Returns the integer code for the pseudoautosomal region.

YchromCode(object): Returns the integer code for the Y chromosome.

MchromCode(object): Returns the integer code for mitochondrial SNPs.

Author(s)

Stephanie Gogarten

See Also

GdsReader, GenotypeData

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
file <- system.file("extdata", "illumina_geno.gds", package="GWASdata")
gds <- GdsGenotypeReader(file)

# dimensions
nsnp(gds)
nscan(gds)

# get snpID and chromosome
snpID <- getSnpID(gds)
chrom <- getChromosome(gds)

# get positions only for chromosome 22
pos22 <- getPosition(gds, index=(chrom == 22))

# get all snps for first scan
geno <- getGenotype(gds, snp=c(1,-1), scan=c(1,1))
length(geno)

# starting at snp 100, get 10 snps for the first 5 scans
getGenotype(gds, snp=c(100,10), scan=c(1,5))

# get snps 1-10, 25-30 for scans 3,5,7
snp.index <- c(1:10, 25:30)
scan.index <- c(3,5,7)
getGenotypeSelection(gds, snp=snp.index, scan=scan.index)

# illustrate drop argument
getGenotypeSelection(gds, snp=5, scan=1:10, drop=TRUE, use.names=FALSE)
getGenotypeSelection(gds, snp=5, scan=1:10, drop=FALSE, use.names=FALSE)

# illustrate order="file" vs order="selection"
snp.index <- c(9,3,5)
scan.index <- c(3,2,1)
getGenotypeSelection(gds, snp=snp.index, scan=scan.index, order="file")
getGenotypeSelection(gds, snp=snp.index, scan=scan.index, order="selection")

close(gds)

GWASTools documentation built on Nov. 8, 2020, 7:49 p.m.