Filter genes

Now that the data have been log-transformed and quantile-normalized, you need to remove the lowly expressed genes that are not relevant to the system being studied.

Den här övningen är en del av kursen

Differential Expression Analysis with limma in R

Visa kurs

Övningsinstruktioner

The ExpressionSet object eset_norm with the normalized Populus data has been loaded in your workspace.

Use plotDensities to visualize the distribution of gene expression levels for each sample. Disable the legend.
Use rowMeans to determine which genes have a mean expression level greater than 5. Name this logical vector keep.
Filter the genes (i.e. rows) of the ExpressionSet object with the logical vector keep and re-visualize.

Interaktiv övning med praktiskt arbete

Testa den här övningen genom att slutföra den här exempelkoden.

library(limma)

# Create new ExpressionSet to store filtered data
eset <- eset_norm

# View the normalized gene expression levels
___(eset, legend = ___); abline(v = 5)

# Determine the genes with mean expression level greater than 5
keep <- ___(exprs(eset)) > ___
sum(keep)

# Filter the genes
eset <- eset[___]
___(eset, legend = ___)

Redigera och kör kod