SeqArray (version 1.12.5)

seqSetFilter-methods: Set a Filter to Sample or Variant

Description

Sets a filter to sample and/or variant.

Usage

"seqSetFilter"(object, variant.sel, sample.sel=NULL, variant.id=NULL, sample.id=NULL, samp.sel=NULL, action=c("set", "intersect", "push", "push+set", "push+intersect", "pop"), verbose=TRUE)
"seqSetFilter"(object, variant.sel, rm.txt="chr", verbose=TRUE)
"seqSetFilter"(object, variant.sel, rm.txt="chr", verbose=TRUE)
"seqSetFilter"(object, variant.sel, chr, verbose=TRUE)
seqResetFilter(object, sample=TRUE, variant=TRUE, verbose=TRUE)
seqSetFilterChrom(object, include=NULL, is.num=NA, from.bp=NULL, to.bp=NULL, verbose=TRUE)

Arguments

object

a SeqVarGDSClass object

variant.sel

a logical/raw/index vector indicating the selected variants; GRanges, a GRanges object for the genomic locations; GRangesList, a GRangesList object for storing a collection of GRanges objects; IRanges, a IRanges object for storing a collection of range objects

sample.sel

a logical/raw/index vector indicating the selected samples

variant.id

ID of selected variants

sample.id

ID of selected samples

samp.sel

it is deprecated since v1.11.12, please use sample.sel instead

action

"set" -- set the current filter via sample.id, variant.id, samp.sel or variant.sel; "intersect" -- set the current filter to the intersection of selected samples and/or variants; "push" -- push the current filter to the stack, and it could be recovered by "pop" later, no change on the current filter; "push+set" -- push the current filter to the stack, and changes the current filter via sample.id, variant.id, samp.sel or variant.sel; "push+intersect" -- push the current filter to the stack, and set the current filter to the intersection of selected samples and/or variants; "pop" -- pop up the last filter

rm.txt

a character, the characters will be removed from seqnames(variant.sel)

chr

a vector of character for chromsome coding

sample

logical, if TRUE, include all samples

variant

logical, if TRUE, include all variants

include

NULL, or character for specified chromosome(s)

is.num

a logical variable: TRUE, chromosome code is numeric; FALSE, chromosome is not numeric; is.num=TRUE is usually used to exclude non-autosomes

from.bp

NULL, no limit; numeric, the lower bound of position

to.bp

NULL, no limit; numeric, the upper bound of position

verbose

if TRUE, show information

Value

None.

Details

seqResetFilter(file) is equivalent to seqSetFilter(file), where the selection arguments in seqSetFilter are NULL.

If from.bp and to.bp has values, they should be equal-size as include. A trio of include, from.bp and to.bp indicates a region on human genomes. NA in from.bp is treated as 0, and NA in to.bp is treated as the maximum of integer (2^31 - 1).

See Also

seqSetFilterChrom, seqGetFilter, seqGetData, seqApply

Examples

# the GDS file
(gds.fn <- seqExampleFileName("gds"))

# display
(f <- seqOpen(gds.fn))

# get 'sample.id
(samp.id <- seqGetData(f, "sample.id"))
# "NA06984" "NA06985" "NA06986" ...

# get 'variant.id'
head(variant.id <- seqGetData(f, "variant.id"))

# get 'chromosome'
table(seqGetData(f, "chromosome"))

# get 'allele'
head(seqGetData(f, "allele"))
# "T,C" "G,A" "G,A" ...


# set sample and variant filters
seqSetFilter(f, sample.id=samp.id[c(2,4,6,8)])
set.seed(100)
seqSetFilter(f, variant.id=sample(variant.id, 5))

# get genotypic data
seqGetData(f, "genotype")


## OR
# set sample and variant filters
seqSetFilter(f, sample.sel=c(2,4,6,8))
set.seed(100)
seqSetFilter(f, variant.sel=sample.int(length(variant.id), 5))

# get genotypic data
seqGetData(f, "genotype")



## set the intersection

seqResetFilter(f)
seqSetFilterChrom(f, 10L)
seqSummary(f, "genotype", check="none")

AF <- seqAlleleFreq(f)
table(AF <= 0.9)

seqSetFilter(f, variant.sel=(AF<=0.9), action="intersect")
seqSummary(f, "genotype", check="none")



## chromosome

seqResetFilter(f)

seqSetFilterChrom(f, is.num=TRUE)
seqSummary(f, "genotype", check="none")

seqSetFilterChrom(f, is.num=FALSE)
seqSummary(f, "genotype", check="none")

seqSetFilterChrom(f, 1:4)
seqSummary(f, "genotype", check="none")
table(seqGetData(f, "chromosome"))

# HLA region
seqSetFilterChrom(f, 6, from.bp=29719561, to.bp=32883508)
seqSummary(f, "genotype", check="none")

# two regions
seqSetFilterChrom(f, c(1, 6), from.bp=c(1000000, 29719561),
    to.bp=c(90000000, 32883508))
seqSummary(f, "genotype", check="none")
seqGetData(f, "chromosome")



# close the GDS file
seqClose(f)