Identify proteins which have too few features to quantify protein abundance in each sample

Protein level abundances are more accurately quantified where there are too more features (PSMs/peptides) to summarise from.

Usually, we are performing the summarisation from a matrix (columns=samples, rows=features) with an associated feature to protein ID mapping. Within the matrix, some values may be missing (NA). In order to correctly identify which proteins can be quantified, we need to start from the feature level object and create a mask which we can use to replace protein-level quantification values with NA. This is what this function does. This can then be combined with the mask_protein_quant function to replace protein level quantification values with NA where they were derived from too few quantification values

Usage

get_protein_no_quant_mask(
  obj,
  min_features,
  master_protein_col = "Master.Protein.Accessions",
  plot = FALSE
)

Arguments

obj

SummarizedExperiment with PSM or peptide-level quantification

min_features

numeric Threshold for minimum features per protein

master_protein_col

character Column name for master protein

plot

Set TRUE to plot how many proteins are quantified in each sample. Horizontal line represents total number of proteins quantified across all samples

Value

Matrix defining whether the protein was quantified from sufficient features