caycheR_scicatch - iPS2-sci-seq perturbation deconvolution

This section describes a typical analysis pipeline similar to the one described in the previous section, but using the following functions: catcheR_scicatch(), catcheR_scicatchQC(), catcheR_filtercatch(), and catcheR_scinocatch().

  1. Prepare the working directory:

    a. Create a subfolder named fastq and copy all the demultiplexed .fastq.gz files (see https://marialuisaratto.github.io/catcheRdocs/catcheR_scicounts.html). Ensure all filenames begin with the well coordinate (e.g., A01).

    b. Copy the cell-by-gene expression matrix CSV file obtained with catcheR_scicount() (see step step-SPFourStepEight of par-SPFourParTwo).

    1. Create a file rc_barcodes_genes.csv with two columns: (1) shRNA barcode; (2) matching shRNA name (format: GENE.shRNAID).

    CAAGAGCC,SMAD2.1
...

    1. Copy the text file sci-RNA-seq-8.RT.oligos used by catcheR_scicount().

  2. Run catcheR_scicatch() to perform a full analysis with automatic thresholding. The arguments are the same as in catcheR_10Xcatch(), except filenames are omitted — files must be present in the fastq folder.

catcheR_scicatch(
    group = c("docker", "sudo"),
    folder,
    expression.matrix,
    reference = "GGCGCGTTCATCTGGGGGAGCCG",
    UCI.length = 6,
    threads = 2,
    percentage = 15,
    ratio = 5,
    mode = "bimodal",
    x = 100,
    y = 400)

Example usage:

catcheR_scicatch(
    group = "docker",
    folder = "path/to/working/folder",
    expression.matrix = "filename.csv",
    threads = 12)

Outputs:

catcheR_scicatch() produces the same key outputs as catcheR_10Xcatch, with the following differences:

  • silencing_matrix.csv contains modified cell names reflecting PCR well and RT barcode:

    P24__RT_27_7_GCCTGTGT_SCR_ACGGTC

    where:

    • P24: PCR well

    • RT_27_7: RT barcode ID

    • GCCTGTGT: shRNA barcode

    • SCR: target gene (e.g. scramble)

    • ACGGTC: UCI

  • Additional QC plots include demux and RT distribution: cell counts per PCR row/column and RT barcode, respectively. These help assess biases during sci-RNA-seq library preparation.