run_jackstraw: Run jackstraw to get genes that are significantly associated...

In scITD: Single-Cell Interpretable Tensor Decomposition

Run jackstraw to get genes that are significantly associated with donor scores for factors extracted by Tucker decomposition

Description

Run jackstraw to get genes that are significantly associated with donor scores for factors extracted by Tucker decomposition

Usage

run_jackstraw(
  container,
  ranks,
  n_fibers = 100,
  n_iter = 500,
  tucker_type = "regular",
  rotation_type = "hybrid",
  seed = container$experiment_params$rand_seed,
  ncores = container$experiment_params$ncores
)

Arguments

container	environment Project container that stores sub-containers for each cell type as well as results and plots from all analyses
ranks	numeric The number of donor ranks and gene ranks to decompose to using Tucker decomposition
n_fibers	numeric The number of fibers the randomly shuffle in each iteration (default=100)
n_iter	numeric The number of shuffling iterations to complete (default=500)
tucker_type	character Set to 'regular' to run regular tucker or to 'sparse' to run tucker with sparsity constraints (default='regular')
rotation_type	character Set to 'hybrid' to perform hybrid rotation on resulting donor factor matrix and loadings. Otherwise set to 'ica_lds' to perform ica rotation on loadings or ica_dsc to perform ica on donor scores. (default='hybrid')
seed	numeric Seed passed to set.seed() (default=container$experiment_params$rand_seed)
ncores	numeric The number of cores to use (default=container$experiment_params$ncores)

Value

The project container with a vector of adjusted pvalues in container$gene_score_associations.

Examples

test_container <- run_jackstraw(test_container, ranks=c(2,4), n_fibers=2, n_iter=10,
tucker_type='regular', rotation_type='hybrid', ncores=1)

scITD documentation built on Sept. 8, 2023, 5:11 p.m.