q_metacell: Quantitative metadata vocabulary for entities

In samWieczorek/Dapar2: Tools for the Differential Analysis of Proteins Abundance with R

Quantitative metadata vocabulary for entities

Description

This function gives the vocabulary used for the quantitative metadata of each entity in each condition.

This function is based on the qMetacell dataframe to look for either missing values (used to update an initial dataset) or imputed values (used when post processing protein qMetacell after aggregation)

In the quantitative columns, a missing value is identified by no value rather than a value equal to 0. Conversion rules Quanti Tag NA or 0 NA

Update the quantitative metadata information of missing values that were imputed

Gives all the tags of the metadata vocabulary containing the pattern (parent and all its children).

Agregation rules for the cells quantitative metadata of peptides. Please refer to the qMetacell.def vocabulary in qMetacell.def()

Usage

metacell.def(level)

custom_metacell_colors()

Set_POV_MEC_tags(conds, df, level)

Metacell_generic(qdata, conds, level)

## S4 method for signature 'SummarizedExperiment'
UpdateMetacellAfterImputation(object, from, to, ...)

search.metacell.tags(pattern, level, depth = "1")

metacombine(met, level)

Arguments

level	xxx
conds	xxx
df	An object of class SummarizedExperiment
qdata	A matrix of quantitative data
object	xxx
from	xxx
to	xxx
...	xxx
pattern	The string to search.
depth	xxx
met	xxx

Value

A data.frame containing the different tags and corresponding colors for the level given in parameter

An instance of class QFeatures.

xxxxx

NA

NA

Glossary

Peptide-level vocabulary

|– 'Any' | | | |– 1.0 'Quantified' | | | | | |– 1.1 "Quant. by direct id" (color 4, white) | | | | | |– 1.2 "Quant. by recovery" (color 3, lightgrey) | | | |– 2.0 "Missing" (no color) | | | | | |– 2.1 "Missing POV" (color 1) | | | | | |– 2.2 'Missing MEC' (color 2) | | | |– 3.0 'Imputed' | | | | | |– 3.1 'Imputed POV' (color 1) | | | | | |– 3.2 'Imputed MEC' (color 2)

Protein-level vocabulary: |– 'Any' | | | |– 1.0 'Quantified' | | | | | |– 1.1 "Quant. by direct id" (color 4, white) | | | | | |– 1.2 "Quant. by recovery" (color 3, lightgrey) | | | |– 2.0 "Missing" | | | | | |– 2.1 "Missing POV" (color 1) | | | | | |– 2.2 'Missing MEC' (color 2) | | | |– 3.0 'Imputed' | | | | | |– 3.1 'Imputed POV' (color 1) | | | | | |– 3.2 'Imputed MEC' (color 2) | | | |– 4.0 'Combined tags' (color 3bis, lightgrey)

Conversion to the glossary

A generic conversion

Conversion for Proline datasets

Conversion from Maxquant datasets

Basic agreagtion

Agregation of non imputed values (2.X) with quantitative values

(1.0, 1.X, 3.0, 3.X)

Not possible

----------------------------

Agregation of different types of missing values (among 2.1, 2.2)

• Agregation of 2.1 peptides between each other gives a missing value non imputed (2.0)

• Agreagtion of 2.2 peptides between each other givesa missing value non imputed (2.0)

• Agregation of a mix of 2.1 and 2.2 gives a missing value non imputed (2.0) |—————————-

Agregation of a mix of quantitative values (among 1.0, 1.1, 1.2, 3.0, 3.X)

• if the type of all the peptides to agregate is 1.0, 1.1 or 1.2, then the final metadata is set the this tag

• if the set of metacell to agregate is a mix of 1.0, 1.1 or 1.2, then the final metadata is set to 1.0

• if the set of metacell to agregate is a mix of 3.X and 3.0, then the final metadata is set to 3.0

• if the set of metacell to agregate is a mix of 3.X and 3.0 and other (X.X), then the final metadata is set to 4.0 |—————————-

Post processing

Update metacell with POV/MEC status for the categories 2.0 and 3.0 TODO

Author(s)

Thomas Burger, Samuel Wieczorek

Samuel Wieczorek

Examples

metacell.def('protein')
metacell.def('peptide')

#-----------------------------------------------
# A shiny app to view color legends 
#-----------------------------------------------
if(interactive()){
data(ft)
ui <- mod_qMetacellLegend_ui("legend")

server <- function(input, output, session) {
  mod_qMetacellLegend_server('legend',
                                object = reactive({ft[[1]]}))
  }
  
 shinyApp(ui = ui, server = server)
  
  
}
data(ft_na)
df <- assay(ft_na, 2)
level <- typeDataset(ft_na, 1)
df <- Set_POV_MEC_tags(ft_na, 1, level)

file <- system.file("extdata", "Exp1_R25_pept.txt", package="DaparToolshedData")
data <- read.table(file, header=TRUE, sep="\t",stringsAsFactors = FALSE)
metadataFile <- system.file("extdata", "samples_Exp1_R25.txt", 
package="DaparToolshedData")
metadata <- read.table(metadataFile, header=TRUE, sep="\t", as.is=TRUE, 
stringsAsFactors = FALSE)
conds <- metadata$Condition
qdata <- data[seq_len(100), seq(56, 61)]
df <- data[seq_len(100) , seq(43,48)]
df <- qMetacell_generic(qdata, conds, 'peptide')


obj <- Exp1_R25_pept[seq_len(10),]
obj[[2]] <- UpdateqMetacell(obj[[2]], 'missing', 'imputed')

search.qMetacell.tags('missing POV', 'peptide')
search.qMetacell.tags('quanti', 'peptide')

## Not run: 
ll <- qMetacell.def('peptide')$node
for (i in 1:length(ll))
test <- lapply(combn(ll, i, simplify = FALSE), 
function(x) tag <- qMetacell_combine(x, 'peptide'))

## End(Not run)

samWieczorek/Dapar2 documentation built on Oct. 12, 2023, 9:54 a.m.