ldtagr {gwascat} | R Documentation |
expand a list of variants by including those in a VCF with LD exceeding some threshold
ldtagr(snprng, tf, samples, genome = "hg19", lbmaf = 0.05, lbR2 = 0.8, radius = 1e+05)
snprng |
a named GRanges for a single SNP. The name must correspond to
the name that will be assigned by |
tf |
TabixFile instance pointing to a bgzipped tabix-indexed VCF file |
samples |
a vector of sample identifiers, if excluded, all samples used |
genome |
tag like 'hg19' |
lbmaf |
lower bound on variant MAF to allow consideration |
lbR2 |
lower bound on R squared for regarding SNP to be incorporated |
radius |
radius of search in bp around the input range |
uses snpStats ld()
a GRanges with names corresponding to 'new' variants and mcols fields 'paramRangeID' (base variant input) and 'R2'
slow but safe approach. probably a matrix method could be substituted using the nice sparse approach already in snpStats
VJ Carey
require(GenomicRanges) cand = GRanges("1", IRanges(113038694, width=1)) names(cand) = "rs883593" require(VariantAnnotation) expath = dir(system.file("vcf", package="gwascat"), patt=".*exon.*gz$", full=TRUE) tf = TabixFile(expath) ldtagr( cand, tf, lbR2 = .8) # should do with 1000 genomes in S3 bucket and gwascat