getVariableLengthData: Get variable-length data

Description Usage Arguments Details Value Author(s) See Also Examples

Description

Get data with multiple values per sample from a GDS object and return as an array

Usage

1
2
## S4 method for signature 'SeqVarGDSClass,character'
getVariableLengthData(gdsobj, var.name, use.names=TRUE, parallel=FALSE)

Arguments

gdsobj

A SeqVarGDSClass object with VCF data.

var.name

Character string with name of the variable, most likely "annotation/format/VARIABLE_NAME".

use.names

A logical indicating whether to assign sample and variant IDs as dimnames of the resulting matrix.

parallel

Logical, numeric, or other value to control parallel processing; see seqParallel for details.

Details

Data which are indicated as having variable length (possibly different numbers of values for each variant) in the VCF header are stored as variable-length data in the GDS file. Each such data object has two components, "length" and "data." "length" indicates how many values there are for each variant, while "data" is a matrix with one row per sample and columns defined as all values for variant 1, followed by all values for variant 2, etc.

getVariableLengthData converts this format to a 3-dimensional array, where the length of the first dimension is the maximum number of values in "length," and the remaining dimensions are sample and variant. Missing values are given as NA. If the first dimension of this array would have length 1, the result is converted to a matrix.

Value

An array with dimensions [n, sample, variant] where n is the maximum number of values possible for a given sample/variant cell. If n=1, a matrix with dimensions [sample,variant].

Author(s)

Stephanie Gogarten

See Also

SeqVarGDSClass, applyMethod, seqGetData

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
file <- system.file("extdata", "gl_chr1.gds", package="SeqVarTools")
gds <- seqOpen(file)
## genotype likelihood 
gl <- seqGetData(gds, "annotation/format/GL")
names(gl)
gl$length
## 3 values per variant - likelihood of RR,RA,AA genotypes
dim(gl$data)
## 85 samples (rows) and 9 variants with 3 values each - 27 columns

gl.array <- getVariableLengthData(gds, "annotation/format/GL")
dim(gl.array)
## 3 genotypes x 85 samples x 9 variants
head(gl.array[1,,])
head(gl.array[2,,])
head(gl.array[3,,])

seqClose(gds)

SeqVarTools documentation built on Nov. 22, 2020, 2 a.m.