Cell_filtering: Filtering perturbed cells with invalid edits.

In bm2-lab/MASCOT: Model-based Understanding of SIngle-cell CRISPR screening

Filtering perturbed cells with invalid edits.

Description

gRNA targets Cas9 to a specific gene locus, but only in 70%-80% will generate true loss-of-function of the targeted gene(Sternberg and Doudna, 2015). Therefore, to estimate the ranking of impact of different perturbation, it is necessary to filter cells with invalid edits.

Usage

Cell_filtering(expression_profile, perturb_information, cpu_num = 4, cell_num_threshold = 30, umi = 0.01, pvalue = 0.05, vargene_min_num = 5, filtered_rate = 0.9, plot = FALSE, plot_path = "./invalid_rate.pdf")

Arguments

expression_profile	A dataframe showing the expression profile after performing the function of "Cell_qc()" and "Data_imputation()".
perturb_information	A character vector showing the information of sample after performing the function of "Cell_qc()" and "Data_imputation()".
cpu_num	The cpu number for parallel computation. The default is 4. Parallel computation is strongly recommeneded to use because this step may take long time without parallel computation.
cell_num_threshold	A cutoff, the minimal perturbed cell number for each perturbation. The default is 30.
umi	The cutoff of average umi to select the differentially expressed genes. The default is 0.01.
pvalue	The p value to select the differentially expressed genes. The default is 0.05.
vargene_min_num	The minimal number of differentially expressed genes. The default is 5. For a perturbation, if the number of differentially expressed genes are less than 5, this perturbation will be filtered directory.
filtered_rate	The default is 0.9. For a specific perturbation, if the influenced cells filtered amount to 90% or higher among all, then such a perturbation was filtered.
plot	FALSE by default. If TRUE, plot the graph to show the ratio of filtered cells for each perturbation.
plot_path	The path of the graph you plot. It works only when the parameter "plot" is TRUE.

Value

expression_profile	The expression profile after performing these filtering steps.
perturb_information	The information (perturbation names and sample names) of cells retained after performing these filtering steps.
perturb_information_abandon	The information (perturbation names and sample names) abandoned after performing these filtering steps.
filter_record	The summary of filtering by these steps.
zero_rate	The proportion of zero expression value in all cells for each perturbation.

Author(s)

Bin Duan

References

1. Sternberg, S.H. & Doudna, J.A. Expanding the Biologist's Toolkit with CRISPR-Cas9. Mol Cell 58, 568-574 (2015). 2. Lappalainen, T. et al. Transcriptome and genome sequencing uncovers functional variation in humans. Nature 501, 506-511 (2013).

bm2-lab/MASCOT documentation built on April 19, 2024, 4:35 p.m.