rraf: Round Robin Allele Frequencies

In poppr: Genetic Analysis of Populations with Mixed Reproduction

Round Robin Allele Frequencies

Description

This function utilizes rrmlg to calculate multilocus genotypes and then subsets each locus by the resulting MLGs to calculate the round-robin allele frequencies used for pgen and psex.

Usage

rraf(gid, pop = NULL, res = "list", by_pop = FALSE, correction = TRUE, ...)

Arguments

gid	a genind or genclone object
pop	either a formula to set the population factor from the strata slot or a vector specifying the population factor for each sample. Defaults to NULL.
res	either "list" (default), "vector", or "data.frame".
by_pop	When this is TRUE, the calculation will be done by population. Defaults to FALSE
correction	a logical indicating whether or not zero-valued allele frequencies should be corrected using the methods outlined in correcting rare alleles. (Default: TRUE)
...	options from correcting rare alleles. The default is to correct allele frequencies to 1/n

Details

Calculating allele frequencies for clonal populations is a difficult task. Frequencies calculated on non-clone-corrected data suffer from bias due to non-independent samples. On the other hand, frequencies calculated on clone-corrected data artificially increases the significance of rare alleles. The method of round-robin allele frequencies as presented in Parks and Werth (1993) provides a method of calculating allele frequencies in a way that minimizes both of these effects.

Rare Alleles

Allele frequencies at a given locus are calculated based on samples that are clone corrected without that locus. When this happens, rare alleles have a high likelihood of dropping out, giving them a frequency of "0". For some analyses, this is a perfectly fine outcome, but for analyses such as pgen and psex, this could result in undefined values. Setting correction = TRUE will allow you to control how these zero-valued allele frequencies are corrected. For details, please see the documentation on correcting rare alleles and examples.

Value

a vector or list of allele frequencies

Note

When by_pop = TRUE, the output will be a matrix of allele frequencies. Additionally, when the argument pop is not NULL, by_pop is automatically TRUE.

Author(s)

Zhian N. Kamvar, Jonah Brooks, Stacy A. Krueger-Hadfield, Erik Sotka

References

Arnaud-Haond, S., Duarte, C. M., Alberto, F., & Serrão, E. A. 2007. Standardizing methods to address clonality in population studies. Molecular Ecology, 16(24), 5115-5139.

Parks, J. C., & Werth, C. R. 1993. A study of spatial features of clones in a population of bracken fern, Pteridium aquilinum (Dennstaedtiaceae). American Journal of Botany, 537-544.

See Also

rrmlg, pgen, psex, rare_allele_correction

Examples

data(Pram)

# Round robin allele frequencies, correcting zero-valued frequencies to 1/nInd(Pram)
rraf(Pram)


## Not run: 

## Round robin allele frequencies will be different than observed

# Compare to without round robin:
PrLoc <- seploc(Pram, res = "mat") # get locus by matrix
lapply(PrLoc, colMeans, na.rm = TRUE)

# Without round robin, clone corrected:
Pcc    <- clonecorrect(Pram, strata = NA) # indiscriminantly clone correct
PccLoc <- seploc(Pcc, res = "mat")
lapply(PccLoc, colMeans, na.rm = TRUE)

## Different methods of obtaining round robin allele frequencies

# Get vector output.
rraf(Pram, res = "vector")

# Getting the output as a data frame allows us to use ggplot2 to visualize
(Prdf <- rraf(Pram, res = "data.frame"))
library("ggplot2")
ggplot(Prdf, aes(y = allele, x = frequency)) +
  geom_point() +
  facet_grid(locus ~ ., scale = "free_y", space = "free")

## Round Robin allele frequencies by population (matrix only)

# By default, allele frequencies will be corrected by 1/n per population
(Prbp <- rraf(Pram, by_pop = TRUE))

# This might be problematic because populations like PistolRSF_OR has a 
# population size of four.

# By using the 'e' argument to rare_allele_correction, this can be set to a
# more reasonable value.
(Prbp <- rraf(Pram, by_pop = TRUE, e = 1/nInd(Pram)))




## End(Not run)

poppr documentation built on May 29, 2024, 5:54 a.m.