Normalization

The first step in any new analysis is to perform normalization. The normalization command performs different actions, depending on the workflow. A Standard Compute runtime is required.

To run normalization, click the Normalize button in the top bar of workbench.

Single cell RNA-seq:

1

Normalize total

Normalizes each cell by total counts over all genes, so that every cell has the same total count after normalization (target = 10,000).

2

Log transform

Takes log1p of normalized counts.

3

Compute highly variable genes

Uses the seurat flavor of the algorithm with a gene threshold of 2,000.

4

Scale counts

Uses a max value of 10.

Bulk RNA-seq

1

Normalize counts

Uses TMM (Trimmed Mean of M component) to make samples comparable.

2

Log transform

Takes log1p of normalized counts.

3

Scale counts

Uses a max value of 10.

For each dataset, Panomics keeps a copy of the raw counts and a copy of the unscaled normalized and log-transformed counts alongside the scaled version.