PCA Overview: The expression data is
log₂(x+1) transformed prior to Principal Component Analysis. This stabilizes variance and prevents genes with extremely high expression from dominating the principal components.
Select to Zoom
Detailed View
Drag to select an area on the thumbnail to zoom in.
Genes are ranked by standard deviations across samples based on the log₂-transformed data.
By showing the patterns of the genes with most variability, this gives us a big picture view of the general pattern of gene expression.