g2_snps: Estimating g2 from larger datasets, such as SNPs

In inbreedR: Analysing Inbreeding Based on Genetic Markers

Description

Estimating g2 from larger datasets, such as SNPs

Usage

g2_snps(genotypes, nperm = 0, nboot = 0, boot_over = "inds", CI = 0.95,
  parallel = FALSE, ncores = NULL, verbose = TRUE)

Arguments

genotypes	data.frame with individuals in rows and loci in columns, containing genotypes coded as 0 (homozygote), 1 (heterozygote) and NA (missing)
nperm	number or permutations for to estimate a p-value
nboot	number of bootstraps to estimate a confidence interval
boot_over	Bootstrap over individuals by specifying "inds" and over loci with "loci". Defaults to "ind".
CI	confidence interval (default to 0.95)
parallel	Default is FALSE. If TRUE, bootstrapping and permutation tests are parallelized
ncores	Specify number of cores to use for parallelization. By default, all available cores are used.
verbose	If FALSE, nothing will be printed to show the status of bootstraps and permutations.

Details

Calculates g2 from SNP datasets. Use convert_raw to convert raw genotypes (with 2 columns per locus) into the required format

Value

g2_snps returns an object of class "inbreed". The functions 'print' and 'plot' are used to print a summary and to plot the distribution of bootstrapped g2 values and CI.

An 'inbreed' object from g2_snps is a list containing the following components:

call	function call.
g2	g2 value
p_val	p value from permutation test
g2_permut	g2 values from permuted genotypes
g2_boot	g2 values from bootstrap samples
CI_boot	confidence interval from bootstrap distribution
se_boot	standard error of g2 from bootstraps
nobs	number of observations
nloc	number of markers

Author(s)

Martin A. Stoffel ([e-mail]) & Mareike Esser ([e-mail])

References

Hoffman, J.I., Simpson, F., David, P., Rijks, J.M., Kuiken, T., Thorne, M.A.S., Lacey, R.C. & Dasmahapatra, K.K. (2014) High-throughput sequencing reveals inbreeding depression in a natural population. Proceedings of the National Academy of Sciences of the United States of America, 111: 3775-3780. Doi: 10.1073/pnas.1318945111

Examples

# load SNP genotypes in 0 (homozygous), 1 (heterozygous), NA (missing) format.
# low number of bootstraps and permutations for computational reasons.
data(mouse_snps)
(g2_mouse <- g2_snps(mouse_snps, nperm = 10, nboot = 10, CI = 0.95, boot_over = "loci"))

# parallelized version for more bootstraps or permutations
## Not run: 
(g2_mouse <- g2_snps(mouse_snps, nperm = 1000, nboot = 1000, 
                     CI = 0.95, parallel = TRUE, ncores = 4))

## End(Not run)

inbreedR documentation built on Feb. 2, 2022, 5:09 p.m.