null.all: Checks for the presence of and determine the frequency of null alleles

Description

The function null.all determines the frequency of null alleles at each locus of a genind object. As an initial step, the function makes a bootstrap estimate (based on the observed allele frequencies) of the probability of seeing the number of homozygotes observed for each allele. If there are a large number of null alleles present at a locus, it would result in multiple alleles at a locus having an excess of homozygotes. The second step of the function estimates the frequency of null alleles and a bootstrap confidence interval for each locus using the methods of Chakraborty et al. (1994) and Brookfield (1996). If the 95% confidence interval includes zero, it indicates that the frequency of null alleles at a locus does not significantly differ from zero.

Usage

null.all(population)

Arguments

population

a genind object (from the package adegenet)

Value

The function returns a list with two main components: homozygotes and null.allele.freq. Homozygotes contains the output of the first part of the analysis: determining the observed number of homozygotes for allele at each locus (homozygotes$observed), generating a distribution of the expected number of homozygotes for each allele at each locus (homozygotes$bootstrap) based upond the observed allele frequencies at a locus, and producing a summary table given the probability of observing the number of homozygotes (homozygotes$probability.obs). null.allele.freq list contains summary tables of the null allele frequency estimates based upon the forumulas of Chakraborty et al. (1994) (summary1), and Brookfield (1996) (summary2). For each summary table, the observed frequency is the null allele frequency determined from the observed heterozygosity and homozygosity at a locus. The median, 2.5th, and 97.5th percentiles are from bootstrap estimates of null allele frequencies obtained by resampling the individual genotypes from the original genind object.Brookfield (1996) provides a brief discussion on which estimator should be used. In summary, it was recommended that Chakraborty et al. (1994)'s method (e.g. summary1) be used if there are individuals with no bands at a locus seen, but they are discounted as possible artefacts. If all individuals have one or more bands at a locus then Brookfield (1996)'s method (e.g. summary2) should be used.

References

Brookfield JFY. (1996) A simple new method for estimating null allele frequency from heterozygote deficiency. Molecular Ecology 5:453-455

Chakraborty R, Zhong Y, Jin L, Budowle B. (1994) Nondetectability of restriction fragments and independence of DNA fragment sizes within and between loci in RFLP typing of DNA. American Journal of Human Genetics 55:391-401

Examples

Run this code

 #not run:
 #data(bilby)
 #here we use only the first 50 individuals to speep up the example
 #popgenreport(bilby, mk.null.all=TRUE, mk.pdf=FALSE)
 
#to get a pdf output you need to have a running Latex version installed on your system.
#popgenreport(bilby, mk.null.all=TRUE, mk.pdf=TRUE)