Skip to contents

Run PCA and store in DimReduction

Source: R/PCA_Functions.R
RunPCA.Rd

Uses generatePCs() under the hood. If HVGs are empty, selects HVGs via FindVariableFeatures(object, nfeatures) with default nfeatures = 2000. If HVGs are present, the PCA uses NFEATURES = length(HVGs(object)). This aligns the PCA feature count with the HVG selection while allowing an explicit override of nfeatures when empty.

Usage

RunPCA(object, nfeatures = 2000)

Arguments

object

A geneda object

nfeatures

Number of features to use when HVGs are empty. Default = 2000

Value

Updated geneda object with DimReduction filled

Examples

data(Snail1KO)
counts <- as.matrix(Snail1KO$rawCounts)
counts <- counts[rowSums(counts >= 10) >= 3, ]
meta <- Snail1KO$metadata
rownames(meta) <- meta$Sample
norm <- log2(t(t(counts) / colSums(counts)) * 1e6 + 1)
obj <- GenEDA(normalized = norm, metadata = meta, counts = counts)
obj <- RunPCA(obj, nfeatures = 1000)
#> HVG slot is empty. Running FindVariableFeatures with top 1000 genes
#> Calculating principal components from top 1000 HVGs
#> Percent variations:
#> PC1: 88.56 %
#> PC2: 5.9 %
#> PC3: 2.91 %
#> PC4: 1.55 %
#> PC5: 1.08 %
#> PC6: 0 %
DimReduction(obj)$percent_var
#>       PC1       PC2       PC3       PC4       PC5       PC6 
#> "88.56 %"   "5.9 %"  "2.91 %"  "1.55 %"  "1.08 %"     "0 %"