scRNA-seq

Single-cell RNA-seq (scRNA-seq) gives inforamtion on the transcripts in individual cells.

Recently, many people have been asking about R specifically so that they can learn how to work with scRNA-seq on their own.

There are several major packages for scRNA-seq: Seurat, a suite of Bioconductor packages, and scanpy. Seurat and Bioconductor are in R, scanpy is Python. They are broadly the same, but are in somewhat of an arms race so some new features or analyses may not be immediately available in all of them.

The book Orchestrating Single-Cell Analysis with Bioconductor is a fantastic, comprehensive resource that even goes through worked examples of published data. This uses the suite of tools in Bioconductor. Highly recommended.

Seurat is another popular package for scRNA-seq, and has a series of vignettes on the home page. There have been recent improvments to the normalization (scTransform v1 and v2), and these use somewhat different steps. The different vignettes therefore differ in the steps, and it can be a bit confusing. The PBMC3k tutorial is the classic starting point, so working through this one will at least give you the context to work through other vignettes.

Note

If you jump right in to the Seurat tutorials without knowing R, you won’t know which commands are standard R and which are Seurat-specific. And if your data do not exactly match their example data, it will be unclear how to modify the code to suit your data if you don’t know R.

You should be fine with the “Level 1” set of R skills above to start using Seurat effectively. See the “Beginner” section above for learning these skills.