ies2xpehh {rehh}

Compute XP-EHH

Description

Compute XP-EHH (standardized ratio of iES of two populations).

Usage

ies2xpehh(
  scan_pop1,
  scan_pop2,
  popname1 = NA,
  popname2 = NA,
  min_nhaplo = NA,
  standardize = TRUE,
  include_freq = FALSE,
  p.side = NA,
  p.adjust.method = "none",
  verbose = TRUE
)

Arguments

scan_pop1

a data frame with markers in rows and columns with chromosome name, position of the marker, frequency of the ancestral allele and iES as obtained by scan_hh on the first population.

scan_pop2

a data frame with markers in rows and columns with chromosome name, position of the marker, frequency of the ancestral allele and iES as obtained by scan_hh on the second population.

popname1

short ID/name of the first population; to be added to an output column name.

popname2

short ID/name of the second population; to be added to an output column name.

min_nhaplo

discard positions where in at least one of the populations fewer than min_nhaplo haplotypes have been evaluated (default NA).

standardize

logical. If TRUE (default), then standardize XP-EHH, else report unstandardized XP-EHH.

include_freq

logical. If TRUE include columns with allele frequencies into result.

p.side

side to which refers the p-value. Default NA, meaning two-sided. Can be set to "left" or "right".

p.adjust.method

method passed to function p.adjust to correct the p-value for multiple testing. Default "none".

verbose

logical. If TRUE (default), report number of markers of the two source data frames and result data frame.

Details

Log ratio of iES (population 1 over population 2) computed as described in Sabeti et al. (2007). Note that the two data frames are merged on the basis of chromosome and position. Marker names are kept, if they are identical and unique in both data frames.

Since the standardized XP-EHH values follow, if markers evolve predominantly neutrally, approximately a standard Gaussian distribution, it is practical to assign to the values a p-value relative to the null-hypothesis of neutral evolution. The parameter p.side determines if the p-value is assigned to both sides of the distribution or to one side of interest.

Value

The returned value is a data frame with markers in rows and columns for chromosome name, marker position, XP-EHH and, if standardized, p-value in a negative log10 scale. Optionally, allele frequencies are included.

References

Gautier, M. and Naves, M. (2011). Footprints of selection in the ancestral admixture of a New World Creole cattle breed. Molecular Ecology, 20, 3128-3143.

Sabeti, P.C. et al. (2007). Genome-wide detection and characterization of positive selection in human populations. Nature, 449, 913-918.

See Also

scan_hh, distribplot, manhattanplot

Examples

library(rehh.data)
data(wgscan.cgu) ; data(wgscan.eut)
## results from a genome scan (44,057 SNPs)
##see ?wgscan.eut and ?wgscan.cgu for details
wgscan.xpehh <- ies2xpehh(wgscan.cgu, wgscan.eut, "CGU", "EUT")