addMissing: Simulate missing values with mimimal assumptions

In TomSmithCGAT/OptProc:

Simulate missing values with mimimal assumptions

Description

Assumes that PSMs from the same peptide should have the same relative abundances, and that the missing values are missing at random (MAR) with peptide intensity the main cause of missingness.

Usage

addMissing(obj, n = Inf, id_column, verbose = TRUE)

Arguments

obj	Input MSnSet
n	Number of PSMs to impute missing values in
id_column	Column to group PSM by
verbose	Output description of missing values to console

Details

Identifies groups of PSMs with at least one with missing values (PSMm) and one without (PSMnm) from the same peptide. PSMnm are then randomly paired with a PSMm, intensity values scaled to the PSMm and missing values (NA) added at the same positions as missing values in PSMm.

Returns two MSnSets: "truth" and "missing". Both have the same dimensions as the input MSnSet. Truth contains the scaled PSMs. Missing contains the scaled PSMs with missing values added.

Value

list("truth":Input data set with ground truths for missing values, "missing":Input data set with NA for missing values)

TomSmithCGAT/OptProc documentation built on July 22, 2023, 12:08 a.m.