generateTSNECoordinates-scRNAseq: generateTSNECoordinates

In ilyessr/conclus: ScRNA-seq Workflow CONCLUS - From CONsensus CLUSters To A Meaningful CONCLUSion

generateTSNECoordinates

Description

The function generates several t-SNE coordinates based on given perplexity and ranges of PCs. The final number of t-SNE plots is length(PCs)*length(perplexities).

Usage

generateTSNECoordinates(theObject, randomSeed=42, cores=2,
                PCs=c(4, 6, 8, 10, 20, 40, 50), perplexities=c(30,40),
                writeOutput = FALSE)

Arguments

theObject	An Object of class scRNASeq for which the count matrix was normalized. See ?normaliseCountMatrix.
randomSeed	Default is 42. Seeds used to generate the tSNE.
cores	Maximum number of jobs that CONCLUS can run in parallel. Default is 1.
PCs	Vector of first principal components. For example, to take ranges 1:5 and 1:10 write c(5, 10). Default = c(4, 6, 8, 10, 20, 40, 50)
perplexities	A vector of perplexity (t-SNE parameter). See details. Default = c(30, 40)
writeOutput	If TRUE, write the tsne parameters to the output directory defined in theObject. Default = FALSE.

Details

Generates an object of fourteen (by default) tables with tSNE coordinates. Fourteen because it will vary seven values of principal components PCs=c(4, 6, 8, 10, 20, 40, 50) and two values of perplexity perplexities=c(30, 40) in all possible combinations. The chosen values of PCs and perplexities can be changed if necessary. We found that this combination works well for sc-RNA-seq datasets with 400-2000 cells. If you have 4000-9000 cells and expect more than 15 clusters, we recommend to take more first PCs and higher perplexity, for example, PCs=c(8, 10, 20, 40, 50, 80, 100) and perplexities=c(200, 240). For details about perplexities parameter see ‘?Rtsne’.

Value

An object of class scRNASeq with its tSNEList slot updated. Also writes coordinates in "dataDirectory/tsnes" subfolder if the parameter writeOutput is TRUE.

Author(s)

Ilyess RACHEDI, based on code by Polina PAVLOVICH and Nicolas DESCOSTES.

See Also

normaliseCountMatrix

Examples

## Object scr containing the results of previous steps
load(system.file("extdata/scrFull.Rdat", package="conclus"))

## Compute the tSNE coordinates
## These parameters are tweaked to fit our example data and reduce
## computing time, please consider using the default parameters or
## adjusted to your dataset.
scr <- generateTSNECoordinates(scr, perplexities=c(2,3),
                                PCs =c(4,5,6,7,8,9,10), cores=2)

ilyessr/conclus documentation built on April 8, 2022, 1:43 p.m.