simulate_data: Simulate differentially linearly coexpressed genes between...

In ryananeff/superNOVA: Differential coexpression within multiple subgroups of data

Simulate differentially linearly coexpressed genes between two groups.

Description

Simulates sets of genes coexpressed with an eigengene which show differential coexpression among two groups. This tool simulates all 9 possible classes of possible gene coexpression changes (positive, negative, null correlation in group A and group B).

Usage

simulate_data(samples_per_group = 50, total_genes = 2000,
  maxCor_true = 0.95, minCor_true = 0.3, maxCor_null = 0.2,
  verbose = T, corr_slope = 1, corr_intercept = 0,
  group1_offset = 0, group2_offset = 0)

Arguments

samples_per_group	The number of samples per group to simulate. Default=50
total_genes	The number of genes among all 9 differential coexpression classes to simulate. Will actually return a number of genes which is a multiple of 9 that is less than this number.
maxCor_true	The absolute value of the maximum correlation rho value for highly correlated pairs.
minCor_true	The absolute value of the minimum correlation rho value for highly correlated pairs.
maxCor_null	The absolute value of the maximum correlation rho value for null correlated pairs. The minimum is 0 correlation.
verbose	Whether or not to print additional information about the simulation when it runs.
corr_slope	The slope of the correlations to simulate. All correlated pairs will have this slope.
corr_intercept	The correlation intercept of the genes in matrix B.
group1_offset	The correlation intercept of group1 (e.g. its mean expression value)
group2_offset	The correlation intercept of group2 (e.g. its mean expression value)

Value

Returns a list of inputs for chowCor, including a matA, matB, design_mat, subgroup names (conditions), and the actual classes of differential coexpression.

ryananeff/superNOVA documentation built on March 29, 2024, 5:31 p.m.