evaluate_performance_deg_calls: evaluate_performance_deg_calls

In alextkalinka/delboy: Differential-representation analysis by Elastic-net Logistic regression with BinOmial-thinning validitY tests

Description

Evaluates the performance of delboy on RNAseq data by comparison with DESeq2 output on the original input data for an experiment, controlling for real signal, and adding known signal (using seqgendiff's binomial-thinning approach) for a sampled number of genes from a logFC distribution with both the number and distribution chosen to match as closely as possible the signal in the real data.

Usage

evaluate_performance_deg_calls(
  data,
  group_1,
  group_2,
  gene_column,
  max.iter,
  num_non_null,
  lfc,
  lfc_dens,
  alpha
)

Arguments

data	A data frame of normalized (and batch-corrected, if necessary) counts for a set of samples with the true signal controlled (by, for example, batch-correcting it).
group_1	A character vector naming the columns that belong to group 1.
group_2	A character vector naming the columns that belong to group 2.
gene_column	A character string naming the column containing gene names.
max.iter	An integer value indicating the maximum number of validation samples.
num_non_null	An integer value indicating the number of genes to add signal to.
lfc	A vector of logFC values for non-null cases.
lfc_dens	A vector of density estimates for the logFC values given in lfc.
alpha	The elastic-net regression penalty, between 0 and 1.

Value

An object of class delboy_performance.

alextkalinka/delboy documentation built on Feb. 2, 2022, 4:19 p.m.