Skip to contents

Calculate v-scores and related statistics for genes in the input counts matrix

get_vscores.Rd

This function calculates v-scores, coefficient of variation (CV), and other statistics for genes in the input counts matrix. The v-score is an above-Poisson noise statistic that helps assess the variability of gene expression levels.

Usage

get_vscores(E, min_mean = 0, nBins = 50, fit_percentile = 0.1, error_wt = 1)

Arguments

E

A counts matrix where rows represent cells and columns represent genes.

min_mean

Minimum mean expression value for genes to be considered.

nBins

Number of bins for calculating running quantiles.

fit_percentile

Percentile used for fitting the running quantile.

error_wt

Weight for the error function during optimization.

Value

A list containing the following components:

  • v_scores: V-scores for each gene.

  • CV_eff: Coefficient of variation (CV) for effective counts.

  • CV_input: Coefficient of variation (CV) for input counts.

  • gene_ix: Indices of genes considered in the analysis.

  • mu_gene: Mean expression values for selected genes.

  • FF_gene: Fano factor (variance to mean ratio) for selected genes.

  • a: Parameter 'a' obtained during optimization.

  • b: Parameter 'b' obtained during optimization.

See also

runningquantile, optimize