Skip to contents

Aggregate Data

aggregate.Rd

Aggregate glycomics or glycoproteomics data to different levels (glycans, glycoforms, glycopeptides, etc.). This function combines quantitative values for each unique combination of specified variables, using sum() by default. It is recommended to call this function after missing value imputation.

The following levels are available:

  • "g": Aggregate glycomics data to glycan compositions.

  • "gs": Aggregate glycomics data to glycan structures.

  • "gf": Aggregate to glycoforms, which is the unique combination of proteins, protein sites, and glycan compositions.

  • "gp": Aggregate to glycopeptides, which is the unique combination of peptides, proteins, protein sites, and glycan compositions.

  • "gfs": Like "gf", but differentiates structures with the same composition.

  • "gps": Like "gp", but differentiates structures with the same composition.

The "g" and "gs" levels are available for glycomics data. The "gf", "gp", "gfs", and "gps" levels are available for glycoproteomics data. When to_level = NULL, glycomics data defaults to "gs" when glycan_structure is present and "g" otherwise. Glycoproteomics data similarly defaults to "gfs" or "gf".

Different levels of aggregation require different columns in the variable information.

  • "g": "glycan_composition"

  • "gs": "glycan_composition", "glycan_structure"

  • "gf": "protein", "glycan_composition", "protein_site"

  • "gp": "peptide", "protein", "glycan_composition", "peptide_site", "protein_site"

  • "gfs": "protein", "glycan_composition", "glycan_structure", "protein_site"

  • "gps": "peptide", "protein", "glycan_composition", "glycan_structure", "peptide_site", "protein_site"

Other columns in the variable information tibble with "many-to-one" relationship with the unique combination of columns listed above will be kept. A common example is the "gene" column. For each "glycoform" (unique combination of "protein", "protein_site", and "glycan_composition"), there should be only one "gene" value, therefore it is kept for "gf" level. On the other hand, the "peptide" column is removed for "gf" level, as one "glycoform" can contain multiple "peptides".

Usage

aggregate(exp, to_level = NULL, standardize_variable = TRUE, f = sum)

# S3 method for class 'glyexp_experiment'
glyclean_aggregate(exp, to_level = NULL, standardize_variable = TRUE, f = sum)

# Default S3 method
glyclean_aggregate(exp, to_level = NULL, standardize_variable = TRUE, f = sum)

Arguments

exp

A glycomics or glycoproteomics container: a glyexp::GlycomicSE(), glyexp::GlycoproteomicSE(), or legacy glyexp_experiment object.

to_level

The aggregation level. If NULL (the default), glycomics data uses "gs" when glycan_structure is present and "g" otherwise; glycoproteomics data uses "gfs" when glycan_structure is present and "gf" otherwise. Explicit values are: "g" (glycan compositions), "gs" (glycan structures), "gf" (glycoforms), "gp" (glycopeptides), "gfs" (glycoforms with structures), or "gps" (glycopeptides with structures). See Details for more information.

standardize_variable

Whether to call glyexp::standardize_variable() after aggregation. Set to FALSE to skip network calls for faster testing. Default is TRUE.

f

A function applied to each group's values separately for each sample. Defaults to sum. Missing values are removed before calling f, which must return one numeric value, including for an empty vector when all values are missing. For example, use mean, max, or function(x) mean(x, trim = 0.1).

Value

A modified container with the same class as exp, an aggregated expression matrix, and updated variable information.