This course introduces single cell RNA-seq data analysis methods, tools and file formats. It covers the preprocessing steps of DropSeq data from raw reads to a digital gene expression matrix (DGE), and how to find sub-populations of cells using clustering with the Seurat tools. You will also learn how to compare two samples and detect conserved cluster markers and differentially expressed genes in them. Both DropSeq and 10X Genomics data are used in the exercises. The user-friendly Chipster software is used in the exercises, so no Unix or R experience is required and the course is thus suitable for everybody. The course takes one day. You will learn how to

• check the quality of reads with FastQC
• tag reads with molecular and cellular barcodes
• trim reads
• align reads to the reference genome with HISAT2 and STAR
• tag reads with gene names
• visualize aligned reads in genomic context using the Chipster genome browser
• estimate the number of usable cells by checking the inflection point
• detect bead synthesis errors
• create and filter DGE
• regress out unwanted variability such as cell cycle affects
• detect variable genes and perform principle component analysis
• cluster cells and find marker genes for a cluster
• run canonical correlation analysis (CCA) to identify common sources of variation between two datasets
• align two samples for integrated analysis
• find conserved cluster markers for two samples
• find differentially expressed genes in a cluster between two samples
• visualize genes with cell type specific responses in two samples

Course material (2018) is available at the course website and it includes:

• slides
• lecture and exercise videos
• exercises. The data is available on Chipster server in the example sessions listed in the exercise sheet, and we also provide ready-made analysis sessions which you can use as a reference when doing exercises on your own. We provide training credentials to Chipster, and you can also log in as guest if you just want to follow the exercises using the ready-made sessions.