Image by Stephanie Hicks via learn.gencore.bio.nyu.edu

Image by 10x Genomics

The 10x library contains four pieces of information, in the form of DNA sequences, for each “read”.

• sample index - identifies the library, with one or two indexes per sample
• 10x barcode - identifies the droplet in the library
• UMI - identifies the transcript molecule within a cell and gene
• insert - the transcript molecule

The sequences for any given fragment will generally be delivered in 3 or 4 files:

• I1: I7 sample index
• I2: I5 sample index if present (dual indexing only)
• R1: 10x barcode + UMI
• R2: insert sequence

The first steps in the analysis of single cell RNAseq data:

• Align reads to genome
  
• Annotate reads with feature (gene)
• Quantify gene expression

• 10x Cell Ranger - This not only carries out the alignment and feature counting, but will also:
  • Call cells
  • Generate a summary report in html format
  • Generate a “cloupe” file

Alternative methods include:

• STAR solo:
  • Generates outputs very similar to CellRanger minus the cloupe file and the QC report
  • Will run with lower memory requirements in a shorter time than Cell Ranger
• Alevin:
  • Based on the popular Salmon tool for bulk RNAseq feature counting
  • Alevin supports both 10x-Chromium and Drop-seq derived data

Setup instructions given in the course materials homepage.

Cell Ranger includes a number of different tools for analysing scRNAseq data, including:

• cellranger mkref - for making custom references
  
• cellranger count - for aligning reads and generating a count matrix
  
• cellranger aggr - for combining multiple samples and normalising the counts

Cell Ranger requires the fastq file names to follow a convention:

<SampleName>_S<SampleNumber>_L00<Lane>_<Read>_001.fastq.gz

e.g. for a single sample in the Caron data set we have:

    SRR9264343_S0_L001_I1_001.fastq.gz
    SRR9264343_S0_L001_R1_001.fastq.gz
    SRR9264343_S0_L001_R2_001.fastq.gz

As with other aligners Cell Ranger requires the information about the genome and transcriptome of interest to be provided in a specific format.

• Obtain from the 10x website for human or mouse (or both - PDX)
• Build a custom reference with cellranger mkref

cellranger mkref \
  --fasta={GENOME FASTA} \
  --genes={ANNOTATION GTF} \
  --genome={OUTPUT FOLDER FOR INDEX} \
  --nthreads={CPUS}

• Computationally very intensive
• High memory requirements

cellranger count \
  --id={OUTPUT_SAMPLE_NAME} \
  --transcriptome={DIRECTORY_WITH_REFERENCE} \
  --fastqs={DIRECTORY_WITH_FASTQ_FILES} \
  --sample={NAME_OF_SAMPLE_IN_FASTQ_FILES} \
  --localcores={NUMBER_OF_CPUS} \
  --localmem={RAM_MEMORY}

• One directory per sample

Two types of outputs:

• Text-based files: .tsv and .mtx
• HDF5 files: .h5

Both of these can be read by standard scRNA-seq analysis packages and contain data for a
unique molecular identified (UMI) count matrix: