sim_add_to_na_flatten: This function overwrites NAs from random indices with "real"...

In mccoy-lab/rhapsodi: R haploid sperm/oocyte data imputation

This function overwrites NAs from random indices with "real" data from the input, in effect sparsifying the input data

Description

This function takes as input a matrix with full "real" data It then creates a vector of all NAs equal in length to the number of cells in the input matrix Then num_nas number of locations are randomly chosen in that vector and overwritten by the corresponding indices from the flattened input data Finally, the vector is reshaped to a matrix matching the size and dimension of the input and returned This function is efficient for higher numbers of NAs (i.e. lower coverage samples)

Usage

sim_add_to_na_flatten(to_add_from, num_nas, num_gametes, num_snps)

Arguments

to_add_from	a matrix, the original data, with nrow of num_snps and ncol of num_gametes
num_nas	an integer, the number of positions that should be randomly chosen to switch to NA
num_gametes	an integer; the number of gametes or the number of columns for to_return
num_snps	an integer; the number of SNPs or the number of rows for to_return

Value

to_return a matrix with nrow of num_snps and ncol of num_gametes. The data is a combination of locations carried over from the input to_add_From and NAs

mccoy-lab/rhapsodi documentation built on July 27, 2022, 3:56 a.m.