cemitool: Full gene co-expression analysis

In csbl-usp/CEMiTool: Co-expression Modules identification Tool

Full gene co-expression analysis

Description

Defines co-expression modules and runs several different analyses.

Usage

cemitool(
  expr,
  annot,
  gmt,
  interactions,
  filter = TRUE,
  filter_pval = 0.1,
  apply_vst = FALSE,
  n_genes,
  eps = 0.1,
  cor_method = c("pearson", "spearman"),
  cor_function = "cor",
  network_type = "unsigned",
  tom_type = "signed",
  set_beta = NULL,
  force_beta = FALSE,
  sample_name_column = "SampleName",
  class_column = "Class",
  merge_similar = TRUE,
  rank_method = "mean",
  ora_pval = 0.05,
  gsea_scale = TRUE,
  gsea_min_size = 15,
  gsea_max_size = 1000,
  min_ngen = 30,
  diss_thresh = 0.8,
  plot = TRUE,
  plot_diagnostics = TRUE,
  order_by_class = TRUE,
  center_func = "mean",
  directed = FALSE,
  verbose = FALSE
)

Arguments

expr	Gene expression data.frame.
annot	Sample annotation data.frame.
gmt	A data.frame containing two columns, one with pathways and one with genes
interactions	A data.frame containing two columns with gene names.
filter	Logical. If TRUE, will filter expression data.
filter_pval	P-value threshold for filtering. Default 0.1.
apply_vst	Logical. If TRUE, will apply Variance Stabilizing Transform before filtering genes. Currently ignored if parameter filter is FALSE.
n_genes	Number of genes left after filtering.
eps	A value for accepted R-squared interval between subsequent beta values. Default is 0.1.
cor_method	A character string indicating which correlation coefficient is to be computed. One of "pearson" or "spearman". Default is "pearson".
cor_function	A character string indicating the correlation function to be used. Supported functions are currently 'cor' and 'bicor'. Default is "cor"
network_type	A character string indicating if network type should be computed as "signed" or "unsigned". Default is "unsigned"
tom_type	A character string indicating if the TOM type should be computed as "signed" or "unsigned". Default is "signed"
set_beta	A value to override the automatically selected beta value. Default is NULL.
force_beta	Whether or not to automatically force a beta value based on number of samples. Default is FALSE.
sample_name_column	A character string indicating the sample column name of the annotation table.
class_column	A character string indicating the class column name of the annotation table.
merge_similar	Logical. If TRUE, merge similar modules.
rank_method	Character string indicating how to rank genes. Either "mean" (the default) or "median".
ora_pval	P-value for overrepresentation analysis. Default 0.05.
gsea_scale	If TRUE, apply z-score transformation for GSEA analysis. Default is TRUE
gsea_min_size	Minimum size of gene sets for GSEA analysis. Default is 15
gsea_max_size	Maximum size of gene sets for GSEA analysis. Default is 1000
min_ngen	Minimal number of genes per submodule. Default 30.
diss_thresh	Module merging correlation threshold for eigengene similarity. Default 0.8.
plot	Logical. If TRUE, plots all figures.
plot_diagnostics	Logical. If TRUE, creates diagnostic plots. Overwritten if plot=FALSE.
order_by_class	Logical. If TRUE, samples in profile plot are ordered by the groups defined by the class_column slot in the sample annotation file. Ignored if there is no sample_annotation file. Default TRUE.
center_func	Character string indicating the centrality measure to show in the plot. Either 'mean' (the default) or 'median'.
directed	Logical. If TRUE, the igraph objects in interactions slot will be directed.
verbose	Logical. If TRUE, reports analysis steps.

Value

Object of class CEMiTool

Examples

# Get example expression data
data(expr0)
# Run CEMiTool analyses
cem <- cemitool(expr=expr0)
# Run CEMiTool applying Variance Stabilizing Transformation to data
cem <- cemitool(expr=expr0, apply_vst=TRUE)
# Run CEMiTool with additional processing messages
cem <- cemitool(expr=expr0, verbose=TRUE)

## Not run: 
# Run full CEMiTool analysis
## Get example sample annotation data
data(sample_annot)
## Read example pathways file
gmt_fname <- system.file("extdata", "pathways.gmt", package = "CEMiTool")
gmt_in <- read_gmt(gmt_fname)
## Get example interactions file
int_df <- read.delim(system.file("extdata", "interactions.tsv", package = "CEMiTool"))
## Run CEMiTool
cem <- cemitool(expr=expr0, annot=sample_annot, gmt=gmt_in,
    interactions=int_df, verbose=TRUE, plot=TRUE)

# Create report as html file
generate_report(cem, directory = "./Report", output_format="html_document")

# Write analysis results into files
write_files(cem, directory="./Tables", force=TRUE)

# Save all plots
save_plots(cem, "all", directory="./Plots")

## End(Not run)

csbl-usp/CEMiTool documentation built on March 27, 2024, 5:14 a.m.