genotype.fsa: Calls alleles for a SNP genotyping experiment

Description

Calls alleles in experiments where fragments of different sizes are expected according to the allele.

genotype.closest.fsa selects for each allele the closest expected normalized peak height among the 0, 1 and 2 copy values provided in the design.

genotype.ratio.fsa calls an allele as absent, heterozygous or homozygous using fixed thresholds on the proportion of signal from a locus.

genotype.N1.fsa calls an allele as present if it exceeds a given proportion of the expected signal for 1 copy, as provided in the design. When a single allele is called present for a locus, it is considered homozygous.

Usage

genotype.closest.fsa(x)
  genotype.ratio.fsa(x, homo = 0.85, hetero = c(0.3, 0.7))
  genotype.N1.fsa(x, threshold = 0.1)

Value

Returns x, with a new or updated genotypes attribute, a data.frame with a row for each locus :

call: The concatenation of the two called alleles, possibly with ?.
alleles: The comma-separated list of ratios observed for all alleles (genotype.ratio.fsa() only).

A calls vector attribute is also set, corresponding to the call column only.

Arguments

x: The fsa object to use, which must have a peaks attribute (see peaks.fsa). Peaks are expected to be named according to the LOCUS_ALLELE pattern. genotype.closest.fsa() expected optional N0, N1 and N2 columns with expected normalized heights for 0, 1 and 2 copies of the allele.
homo: Single numeric value, the ratio of the considered allele signal over all signal from this locus to call a homozygous allele. Similarly alleles below 1 - homo will be called "absent".
hetero: Numeric vector of length two, minimum and maximum ratios of the considered allele signal over all signal from this locus to call a heterozygous allele.
threshold: Single numeric value, the ratio of the considered allele signal over the expected value for 1n to call an allele.

Author

Sylvain Mareschal

Examples

Run this code

  # Using a design file
  design <- designFile(system.file("extdata/design_SNP.conf", package="FSAtools"))
  
  # Example FSA file provided
  fsa <- read.fsa(system.file("extdata/fsa_SNP/A7840.fsa", package="FSAtools"))
  fsa <- align.fsa(
    fsa,
    channel = design$align.fsa$channel,
    outThreshold = design$align.fsa$outThreshold,
    useLadder = design$align.fsa$useLadder
  )
  fsa <- peaks.fsa(fsa, peaks=design$GLOBALS$PEAKS)
  
  # Genotype
  fsa <- genotype.ratio.fsa(fsa)
  print(attr(fsa, "genotypes"))
  print(attr(fsa, "calls"))