getHapmapProbabilities {VanillaICE}R Documentation

Estimate probability of correct genotype call

Description

This function is used by the ICE HMM described in the reference below and currently only supported for 50k and 250k Affymetrix SNP chips. A nonparametric estimate of the probability that the genotype call is correct is based on the distribution of crlmm confidence scores in the reference HapMap population, where we assume that we know the true genotype. A similar approach could be used to estimate probabilities with other genotyping algorithms. See reference below for more information.

Usage

getHapmapProbabilities(object)

Arguments

object One of the HMM classes defined in ICE

Details

The annotation slot in the object must contain one of the following strings that identifies the Affymetrix chip:

mapping100k (if the 50k xba and 50k hind results have been merged into one object)

pd.mapping50k.xba240 (50k Xba platform)

pd.mapping50k.hind240 (50k Hind)

mapping500k (if the two 250k chips – Nsp and Sty – have been merged)

pd.mapping250k.nsp

pd.mapping250k.sty

Value

A vector of probabilities that the genotype call is correct.

Note

Estimated probabilities are chip-specific. This function is typically not called directly.

Author(s)

References

RB Scharpf et al. (2007), Department of Biostatistics Working Papers, Working Paper 136

See Also

callEmission, crlmm

Examples






[Package VanillaICE version 1.0.2 Index]