extractTransitions: Extract scores for given cell transitions

Description Usage Arguments Value See Also Examples

View source: R/extractTransitions.R

Description

This function extracts the values of the CellScore for all the test samples of a given set of (valid) cell transition. While it can be used as standalone, it serves as an internal function for several other CellScore functions.

Usage

1
extractTransitions(cellscore, cell.change)

Arguments

cellscore

a data.frame of CellScore values as calculated by the function CellScore().

cell.change

a data frame containing three columns, one for the start (donor) test and target cell type. Each row of the data. frame describes one transition from the start to a target cell type.

Value

This function returns a data frame with the same columns as the input data frame cellscore, extended with additional column that is used as a single identifier of each valid cell transition. Technically, the output is subselection of the input data frame.

See Also

CellScore for details on CellScore calcualtion.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
## Load the expression set for the standard cell types
library(Biobase)
library(hgu133plus2CellScore) # eset.std

## Locate the external data files in the CellScore package
rdata.path <- system.file("extdata", "eset48.RData", package = "CellScore")
tsvdata.path <- system.file("extdata", "cell_change_test.tsv",
                             package = "CellScore")

if (file.exists(rdata.path) && file.exists(tsvdata.path)) {

   ## Load the expression set with normalized expressions of 48 test samples
   load(rdata.path)

   ## Import the cell change info for the loaded test samples
   cell.change <- read.delim(file= tsvdata.path, sep="\t",
                             header=TRUE, stringsAsFactors=FALSE)

   ## Combine the standards and the test data
   eset <- combine(eset.std, eset48)

   ## Generate cosine similarity for the combined data
   ## NOTE: May take 1-2 minutes on the full eset object
   ## so we subset it for 4 cell types
   pdata <- pData(eset)
   sel.samples <- pdata$general_cell_type %in% c("ESC", "EC", "FIB", "KER", 
                 "ASC", "NPC", "MSC", "iPS", "piPS")
   eset.sub <- eset[, sel.samples]
   cs <- CosineSimScore(eset.sub, cell.change, iqr.cutoff=0.1)

   ## Generate the on/off scores for the combined data
   individ.OnOff <- OnOff(eset.sub, cell.change, out.put="individual")

   ## Generate the CellScore values for all samples
   cellscore <- CellScore(eset.sub, cell.change, individ.OnOff$scores,
                          cs$cosine.samples)

   ## Extract the scores for the transitions given in cell.change
   cellscore.cc <- extractTransitions(cellscore, cell.change)

   ## View the sub_cell_type1 in the extracted object, it should be the same
   ## as the test cell types named in cell.change
   table(cellscore.cc$sub_cell_type1)
}

CellScore documentation built on Nov. 1, 2018, 3:48 a.m.