calc_p_from_bootstraps: Calculate p-values from bootstrapped distributions.

In hemstrow/snpR: Whole-Genome Analysis Tools for Use with Single Nucleotide Polymorphism Data

Calculate p-values from bootstrapped distributions.

Description

calc_p_from_bootstraps finds p-values for observed smoothed window statistics from bootstrapped distributions, such as produced by do_bootstraps.

Usage

calc_p_from_bootstraps(
  x,
  facets = "all",
  statistics = "all",
  alt = "two-sided",
  par = FALSE,
  fwe_method = "BY",
  fwe_case = c("by_facet", "overall")
)

Arguments

x	snpRdata object.
facets	character, default "all". Facets to use.
statistics	character, default "all". Vector naming the statistics to calculate p-values for. By default calculates p-values for all possible stats.
alt	character, default "two-sided". Specifies the alternative hypothesis to be used. Options: "less": probability that a bootstrapped value is as small or smaller than observed. "greater": probability that a bootstrapped value is as large or larger than observed. "two-sided": probability that a bootstrapped value is as or more extreme than observed.
par	numeric or FALSE, default FALSE. If numeric, the number of cores to use for parallel processing.
fwe_method	character, default c("bonferroni", "holm", "BH", "BY"). Type of Family-Wise Error correction (multiple testing correction) to use. For details and options, see p.adjust.
fwe_case	character, default c("by_facet", "by_subfacet", "overall"). How should Family-Wise Error correction (multiple testing correction) be applied? "by_facet": Each facet supplied (such as pop or pop.fam) is treated as a set of tests. "by_subfacet": Each level of each subfacet is treated as a separate set of tests. "overall": All tests are treated as a set.

Details

Calculates p-values for smoothed values of a statistic based upon a bootstrapped null distribution of that statistic using an empirical continuous distribution function.

p-values can be generated for specific snp or sample metadata categories using the facets argument, as described in Facets_in_snpR. Only facets for which bootstrap data and raw statistical data have both been calculated will be run. "all" and NULL follow the typical facet rules.

Likewise, p-values can be generated for specific statistics using the statistics argument. Only statistics for which bootstrap data and raw statistical data have both been calculated will be run. By default, all stats for which a bootstrap null distribution has been generated will be run.

Value

snpRdata object, with p-values merged into the stats or pairwise.stats sockets.

Author(s)

William Hemstrom

See Also

ecdf

Examples

## Not run: 
# add statistics and generate bootstraps
x <- calc_basic_snp_stats(stickSNPs, c("chr.pop"), sigma = 200, step = 150)
x <- do_bootstraps(x, facets = c("chr.pop"), boots = 1000, sigma = 200, step = 150)
x <- calc_p_from_bootstraps(x)
get.snpR.stats(x, "chr.pop", "single.window") # pi, ho, etc
get.snpR.stats(x, "chr.pop", "pairwise.window") # fst

## End(Not run)

hemstrow/snpR documentation built on March 20, 2024, 7:03 a.m.