Skip to content

Latest commit

 

History

History
72 lines (56 loc) · 3.54 KB

usage_preworkflow.md

File metadata and controls

72 lines (56 loc) · 3.54 KB

Pre-Workflow: Demultiplexing with hashtag oligos

Running the script

The Pre-Workflow can be run from outside the actual Rmarkdown script. When you are using the render function of the rmarkdown package you can run the script as follows:

rmarkdown::render(
    "scrnaseq_hto.Rmd",
    output_format = "html_document",
    output_dir = ".",
    output_file = "test",
    params = paramsList)

with paramsList set as:

paramsList = list()

paramsList$project = "HTO_testDataset"
paramsList$path_data = "test_datasets/10x_pbmc_hto_GSE108313/counts"
paramsList$path_out = "test_datasets/10x_pbmc_hto_GSE108313/demultiplexed"
paramsList$hto_names = setNames(c("htoA","htoB","htoC","htoD","htoE","htoF","htoG","htoH"), c("htoA","htoB","htoC","htoD","htoE","htoF","htoG","htoH"))
paramsList$mt = "^MT-"
paramsList$col = "palevioletred"
paramsList$sample_cells = NULL

Arguments

project

ID of the project (Default: "HTO_testDataset")

path_data

Input directory where data are located (Default: "test_datasets/10x_pbmc_hto_GSE108313/counts")

stats

Path to a mapping statistics file, e.g. "metrics_summary.csv" generated by Cell Ranger, but can be any table or NA if not available (Default: NA)

path_out

Output directory where the results will be saved (Default: "test_datasets/10x_pbmc_hto_GSE108313/demultiplexed")

hto_names

HTOs have an ID that is included in the "features.tsv" input file. We additionally ask for readable names that are used throughout the report. Names could look as follows, where HTO1-3 are the IDs included in raw dataset: param$hto_names = setNames(c("NameA", "NameB", "NameC"), c("HTO1", "HTO2", "HTO3")) (Default: c("htoA", "htoB", "htoC", "htoD", "htoE", "htoF", "htoG", "htoH"), c("htoA", "htoB", "htoC", "htoD", "htoE", "htoF", "htoG", "htoH"))

hto_regex

Alternatively, a regular expression can be provided that finds the HTO ids included in the "features.tsv" input file. The hto_names argument always supersedes the hto_regex argument so be sure to set it to NULL or comment it out.
param$hto_regex = "^hto" (Not used by default.)

norm

Normalisation method used for the HTO counts. Default is the LogNormalize method of Seurat where the counts of each cell are divided by the total counts for that cell, multiplied by 10,000 and then natural-log transformed. Alternatively, the CLR (centered log ratio) method can be used where the counts of each cell are divided by the geometric mean of the counts for that cell and then natural-log transformed. (Default: LogNormalize)

mt

Prefix of mitochondrial genes (Default: "^MT-")

col

Main colour(s) to use for plots (Defaults: "palevioletred")

sample_cells

Sample data to at most n cells (mainly for tests); set to NULL to deactivate (Default: NULL)

path_to_git

Path to to this git repository (Default: ".")

default_debugging

This sets the debugging behaviour in case of errors. When running the script in RStudio, set to default_debugging to use the default interactive debugger. When running the script on linux system without X11 support, set to terminal_debugger which will then start the debugger in the terminal. For running the script non-interactively (e.g. on a cluster), set to print_traceback which will deactivate any interactive debugger and will just print a traceback of the error. (Default: default_debugging