Perform Gene-set Enrichment Analyses Using SuSiE

perform_gsea(
  gene_signals,
  gene_sets,
  gene_set_info = NULL,
  min_size = 10,
  max_size = 400,
  top_genes = 10,
  L = 10,
  coverage = 0.95,
  max_iter = 100,
  tol = 0.001,
  verbose = TRUE
)

Arguments

gene_signals: A matrix of signals in which rows correspond to genes and columns correspond to the different signals. If there is only one signal, this can be a vector.
gene_sets: A binary matrix (sparse or dense) of gene sets in which rows correspond to genes and columns correspond to gene sets. An entry of 1 means the gene is a member of the gene set.
gene_set_info: Optional data frame containing additional information about the gene sets.
min_size: Gene sets with fewer genes than this will not be considered.
max_size: Gene sets with more genes than this will not be considered.
top_genes: The number of “top genes” to include in the
L: The maximum number of selected gene sets; passed as the “L” argument to susie.
coverage: The “coverage” parameter in susie.
max_iter: The “max_iter” parameter in susie.
tol: The “tol” parameter in susie.
verbose: If verbose = TRUE, print updates about progress of the analysis.

Value

A list containing (a) the SuSiE model fits (one for each signal); (b) a tibble containing the gene set enrichment results.

Examples


# Add an example here illustrating the use of perform_gsea().
library(Matrix)
library(pathways)
#> 
#> Attaching package: ‘pathways’
#> The following object is masked from ‘package:singlecelljamboreeR’:
#> 
#>     perform_gsea
set.seed(1)
data(gene_sets_human)
i <- which(!is.na(gene_sets_human$gene_info$Ensembl))
y <- gene_sets_human$gene_sets[i,"M973"] + 0.1 * rnorm(length(i))
out <- singlecelljamboreeR::perform_gsea(y,gene_sets_human$gene_sets[i,])
#> Number of gene signals: 1 
#> Number of gene sets: 28649 
#> Number of genes: 34954 
#> signal k1:
#> [1] "objective:30273.0903457171"
#> [1] "objective:30551.4920829527"
#> [1] "objective:30734.0400488999"
#> [1] "objective:30734.0428220968"
#> [1] "objective:30737.3019186511"
#> [1] "objective:30737.3020603809"
#> [1] "objective:30743.8271067539"
#> [1] "objective:30743.8274125568"
#> [1] "objective:30743.845098598"
#> [1] "objective:30743.8450986074"
#> [1] "objective:30743.8451031422"
#> Compiling the results.
out$selected_gene_sets
#> # A tibble: 1 × 8
#>   signal CS    gene_set   lbf   pip  coef genes top_genes 
#>   <fct>  <fct> <chr>    <dbl> <dbl> <dbl> <dbl> <list>    
#> 1 k1     L1    M973     3504.     1  1.01    70 <chr [10]>