5. Visualization

Scaling the data

We continue the Seurat tutorial on the analysis of Peripheral Blood Mononuclear Cells (PBMC) at the Scaling the data step.

Why do we need to scale the data before performing a PCA ?

Perform linear dimensional reduction

What is the first information that you need to check from a PCA computation ?

Why to the heatmap look noisier as the PC number increases ?

Determine the ‘dimensionality’ of the dataset

Compare the number of PC selected with the number of genes we are working with.

Cluster the cells

Why do we work with a KNN graph for the clustering instead of working with the full adjacency matrix ?

Run non-linear dimensional reduction (UMAP/tSNE)

For the UMAP projection what can you say about the position of the cluster 3 compared to the cluster 1 ?

Compare the UMAP projection with t-SNE

In the next section where you will learn how to perform differential analyses on scRNASeq data.