sliding.windows.fstat: Compute sliding window estimates of F-statistics or ratio of F-statistics over the genome

Description

Compute sliding window estimates of F-statistics or ratio of F-statistics over the genome

Usage

sliding.windows.fstat(
  x,
  num.pop.idx = NULL,
  den.pop.idx = NULL,
  num.stat = NULL,
  den.stat = NULL,
  window.def = c("nsnp", "bp")[1],
  sliding.window.size = NULL,
  window.overlap.fact = 0.5,
  bp.start.first.snp = TRUE,
  verbose = TRUE
)

Value

A data frame with 7 columns with for each window in a row their i) chromosome/contig of origin; ii) start and iii) end position; iv) the mid-position of each window; v) the cumulated mid-position of each window (to facilitate further plotting); vi) the number of SNPs included in the computation of window value; and vii) the estimated value of the statistic

Arguments

x: A pooldata object containing Pool-Seq information or a countdata object containing allele count information
num.pop.idx: A vector of length 1 to 4 (depending on num.stat) giving the index of the populations. If num.stat="het", num.pop.idx must be of length 1: num.pop.idx=i specifies the ith pop in x. If num.stat="div", "F2" or "Fst", num.pop.idx must be of length 2: num.pop.idx=c(i,j) specifies the pairs of populations with indexes i and j in x. If num.stat="F3" or "F3star", num.pop.idx must be of length 3 (num.pop.idx=c(i,j,k) specifies the F3(pop_i;pop_j,pop_k) populations triplet). Finally, if num.stat="F4" or "Dstat", num.pop.idx must be of length 4: num.pop.idx=c(i,j,k) specifies the F4(pop_i,pop_j;pop_k,pop_l) populations quadruplet i.e. the computed (numerator) statistic computed is (F2(pop_i,pop_k)-F2(pop_i,pop_l)-F2(pop_j,pop_k)+F2(pop_j,pop_l))/2.
den.pop.idx: A vector of length 1 to 4 (see num.pop.idx description) giving the index of the populations specifying the F-statistic. If NULL, the computed statistic is the one specified by num.pop.idx.
num.stat: the name of the (numerator) stat which must be "het" (1-Q1), "div" (1-Q2), "F2", "Fst", "F3", "F3star", "F4" or "Dstat"
den.stat: the name of the (numerator) stat which must be "het" (1-Q1), "div" (1-Q2), "F2", "Fst", "F3", "F3star", "F4" or "Dstat"
window.def: Either "nsnp" or "bp" to define windows by either a number of SNPs or a size in bp, respectively
sliding.window.size: A numeric value giving the number of SNPs or the size (in bp) of the windows depending window.def
window.overlap.fact: A numeric value (between 0 and 1) giving the percentage of overlap between consecutive windows (default=0.5)
bp.start.first.snp: When window.def="bp", if TRUE (default) the windowing start at the first SNP position, if FALSE the windowing start at position 1
verbose: If TRUE extra information is printed on the terminal

Details

Compute sliding window estimates of F-statistics or ratio of F-statistics over the genome.

Examples

Run this code

 make.example.files(writing.dir=tempdir())
 pooldata=popsync2pooldata(sync.file=paste0(tempdir(),"/ex.sync.gz"),poolsizes=rep(50,15))