calc_sfs: Generate a 1-2d site frequency spectrum from a snpRdata...

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

Generate a 1-2d site frequency spectrum from a snpRdata object.

Description

Generates a 1 or 2 dimensional site frequency spectrum from a dadi input file using the projection methods and folding methods of Marth et al (2004) and Gutenkunst et al (2009). This code is essentially an R re-implementation of the SFS construction methods implemented in the program dadi (see Gutenkunst et al (2009)).

Usage

calc_sfs(
  x,
  facet = NULL,
  pops = NULL,
  projection,
  fold = TRUE,
  update_bib = FALSE
)

Arguments

x	snpRdata object. The SNP metadata must contain "ref" and "anc" data.
facet	character, default NULL. Name of the sample metadata column which specifies the source population of individuals. For now, allows only a single simple facet (one column).If NULL, runs the entire dataset.
pops	character, default NULL. A vector of population names of up to length 2 containing the names of populations for which the an SFS is to be created. If NULL, runs the entire dataset.
projection	numeric. A vector of sample sizes to project the SFS to, in number of gene copies. Sizes too large will result in a SFS containing few or no SNPs. Must match the length of the provided pops vector.
fold	logical, default FALSE. Determines if the SFS should be folded or left polarized. If FALSE, snp metadata columns named "ref" and "anc" containing the identity of the derived and ancestral alleles, respectively, should be present for polarization to be meaningful.
update_bib	character or FALSE, default FALSE. If a file path to an existing .bib library or to a valid path for a new one, will update or create a .bib file including any new citations for methods used. Useful given that this function does not return a snpRdata object, so a citations cannot be used to fetch references.

Details

Site frequency spectrums are constructed using the projection methods detailed in Marth et al (2004) and the 2 dimensional expansion in Gutenkunst et al (2009). Folding methods are also taken from Gutenkunst et al (2009). Either 1 or 2d SFSs can be constructed by providing a vector of population names and projection sizes.

Note that ref and anc columns are suggested in the SNP metadata, containing the derived and ancestral character states, respectively. These should contain three characters each: two flanking bases and the SNP. For example, for an A/C SNP flanked by a G and a T, "GCT" and "GAT" would be expected. Note that if these character states are not known, the minor and major alleles will be substituted. Unfolded spectra will be misleading in this case.

Value

A matrix or vector containing the site frequency spectrum with a "pops" attribute containing population IDs, such as c("POP1", "POP2"). For a 2d SFS, the first pop is the matrix columns and the second is the matrix rows.

Author(s)

William Hemstrom

References

Gutenkunst et al (2009). Inferring the joint demographic history of multiple populations from multidimensional SNP frequency data. PLoS genetics, 5(10), e1000695.

Marth et al (2004). The allele frequency spectrum in genome-wide human variation data reveals signals of differential demographic history in three large world populations. Genetics, 166(1), 351-372.

Examples

## Not run: 
# add the needed ref and anc columns, using the major and minor alleles (will fold later)
dat <- calc_maf(stickSNPs)
# note, setting ref and anc is done by default if these columns don't exist!
snp.meta(dat)$ref <- paste0("A", get.snpR.stats(dat)$minor, "A") 
snp.meta(dat)$anc <- paste0("A", get.snpR.stats(dat)$major, "A")

# run for two populations
## call calc_sfs()
sfs <- calc_sfs(dat, "pop", c("ASP", "CLF"), c(10,10))
## plot
plot_sfs(x = sfs)


# run for the overall dataset
sfs <- calc_sfs(dat, projection = 30)
## plot
plot_sfs(x = sfs)

# note that plot_sfs() will take a snpRdata object, calling calc_sfs()
plot_sfs(dat, projection = 30)

## End(Not run)

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