Import gmt files

Gmt files can be imported thanks to the format_from_gmt function. Let’s use saved data from the package:

# Retrieve path of the example gmt
file<-system.file("extdata", "cellcycle_short.gmt", package = "ACSNMineR")
# Then import it
gmt<-ACSNMineR::format_from_gmt(file)

ACSN maps are built-in and can easily be accessed through ACSNEnrcihment::ACSN_maps:

# Name of available maps:
names(ACSNMineR::ACSN_maps)

#And accessing them:
ACSNMineR::ACSN_maps$CellCycle

The gene set that was used for tests is the following:

ACSNMineR::genes_test

Perform analysis

Gene set enrichment for a single set can be performed by calling:

Example<-ACSNMineR::enrichment(ACSNMineR::genes_test,
    min_module_size = 10, 
    threshold = 0.05,
    maps = list(cellcycle = ACSNMineR::ACSN_maps$CellCycle))

Where:

• genes_test is a character vector to test

• min_module_size is the minimal size of a module to be taken into account

• threshold is the maximal p-value that will be displayed in the results (all modules with p-values higher than threshold will be removed)

• maps is a list of maps -here we take the cell cycle map from ACSN- imported through the format_from_gmt() function of the package

Gene set enrichment for multiple sets/cohorts can be performed by calling:

Example<-ACSNMineR::multisample_enrichment(Genes_by_sample = list(set1 = ACSNMineR::genes_test[-1],
                                                              set2 = ACSNMineR::genes_test[-2]),
    maps = ACSNMineR::ACSN_maps$CellCycle,
    min_module_size = 10,
    cohort_threshold = FALSE)

print(Example[[1]])

print(Example[[2]])

Where:

• Genes_by_sample is a list of character vectors to test

• min_module_size is the minimal size of a module to be taken into account

• maps is a list of maps -here we take the cell cycle map from ACSN - imported through the format_from_gmt() function of the package

Heatmaps

Heatmaps for single sample or multiple sample representing p-values can be easily generated thanks to the represent_enrichment function.

ACSNMineR::represent_enrichment(enrichment = list(
    SampleA = ACSNMineR::enrichment_test[1:10,], 
    SampleB = ACSNMineR::enrichment_test[3:10,]),
    plot = "heatmap", 
    scale = "reverselog",
    low = "steelblue" , high ="white",
    na.value = "grey")+theme(axis.text = element_text(size = 6,angle = 45),
                             legend.text = element_text(size = 6),
                             legend.title = element_text(size = 8))

Where:

• enrichment is the result from the enrichment or multisample_enrichment function

• scale can be set to either identity or log and will affect the gradient of colors

• low: the color for the low (significant) p-values

• high: color for the high (less significant) p-values

• na.value is the color in which tiles which have “NA” should appear

Barplots

A barplot can be achieved by using the following:

ACSNMineR::represent_enrichment(enrichment = list(
    SampleA = ACSNMineR::enrichment_test[1:10,], 
    SampleB = ACSNMineR::enrichment_test[3:10,]),
    plot = "bar", 
    scale = "reverselog")

Where:

• enrichment is the result from the enrichment or multisample_enrichment function

• scale can be set to either identity or log and will affect the gradient of colors

• nrow is the number of rows that should be used to plot all barplots (default is 1)