predict.sex: Predict sample sexes based on genotype and intensity data

In andrewparkermorgan/argyle: Basic interface and QC for genotypes from Illumina Infinium arrays

Description

Predict sample sexes based on genotype and intensity data

Usage

## S3 method for class 'sex'
predict(object, method = c("ycalls", "xy"), clean = TRUE,
  platform = "giga", ...)

Arguments

object	a genotypes object
method	how to go about making sex predictions (see Details)
clean	logical; for *MUGA arrays, use only known-good Y chromosome markers
platform	character; name of a specific array platform (used only for "xy" method)
...	other parameters passed to underlying prediction functions

Details

Implements several (soon) methods for predicting the sex of a sample given genotype calls and hybridization intensity on the sex chromosomes. Assumes the sex-chromosome system of eutherian mammals (female karyotype XX, male karyotype XY). Sex chromosomes should be named "X" ("chrX") and "Y" ("chrY") respectively.

Method "ycalls" simply counts the number of missing and of non-missing, non-heterozygous calls at Y-linked markers, and applies a threshold to both. The defaults are calibrated to the GigaMUGA array for mouse. Females have mostly missing calls, males have mostly non-missing calls, and no sample should have many heterozygous calls.

Method "xy" calculates the sum-intensity on each sex chromsome and evaluates them against a pre-constructed set of clusters in 2d space corresponding to each sex. Only available for the GigaMUGA array for mouse, at present.

Value

a dataframe with 4 columns: individual ID, nominal sex (0 if unknown), predicted sex (0 if ambiguous), probability (NA if not a model-based prediction)

andrewparkermorgan/argyle documentation built on May 10, 2019, 11:08 a.m.