gen_results: Generate results

In neurogenomics/MultiEWCE: Multi-Scale Target Explorer

Generate results

Description

Generates EWCE results on multiple gene lists in parallel by calling ewce_para. It allows you to stop the analysis and then continue later from where you left off as it checks the results output directory for finished gene lists and removes them from the input. It also excludes gene lists with less than 4 unique genes (which cause errors in EWCE analysis).

Usage

gen_results(
  ctd,
  gene_data,
  list_name_column = "hpo_id",
  gene_column = "gene_symbol",
  list_names = unique(gene_data[[list_name_column]]),
  reps = 100,
  annotLevel = 1,
  genelistSpecies = "human",
  sctSpecies = "human",
  force_new = FALSE,
  bg = get_bg(species1 = genelistSpecies, species2 = sctSpecies, overwrite = force_new),
  min_genes = 4,
  cores = 1,
  parallel_boot = FALSE,
  save_dir_tmp = NULL,
  save_dir = tempdir(),
  verbose = 1,
  ...
)

Arguments

ctd	CellTypeDataset generated using generate_celltype_data.
gene_data	data frame of gene list names and genes (see get_gene_lists).
list_name_column	The name of the gene_data column that has the gene list names.
gene_column	The name of the gene_data column that contains the genes.
list_names	character vector of gene list names.
reps	Number of random gene lists to generate (Default: 100, but should be >=10,000 for publication-quality results).
annotLevel	An integer indicating which level of sct_data to analyse (Default: 1).
genelistSpecies	Species that hits genes came from (no longer limited to just "mouse" and "human"). See list_species for all available species.
sctSpecies	Species that sct_data is currently formatted as (no longer limited to just "mouse" and "human"). See list_species for all available species.
force_new	Overwrite previous results in the save_dir_tmp.
bg	List of gene symbols containing the background gene list (including hit genes). If bg=NULL, an appropriate gene background will be created automatically.
min_genes	Minimum number of genes per list (default: 4)
cores	The number of cores to run in parallel (e.g. 8) int.
parallel_boot	Parallelise at the level of bootstrap iterations, rather than across gene lists.
save_dir_tmp	Folder to save intermediate results files to (one file per gene list). Set to NULL to skip saving temporary files.
save_dir	Folder to save merged results in.
verbose	Print messages.
...	Arguments passed on to EWCE::bootstrap_enrichment_test sct_data List generated using generate_celltype_data. hits List of gene symbols containing the target gene list. Will automatically be converted to human gene symbols if geneSizeControl=TRUE. sctSpecies_origin Species that the sct_data originally came from, regardless of its current gene format (e.g. it was previously converted from mouse to human gene orthologs). This is used for computing an appropriate backgrund. output_species Species to convert sct_data and hits to (Default: "human"). See list_species for all available species. method R package to use for gene mapping: "gprofiler" : Slower but more species and genes. "homologene" : Faster but fewer species and genes. "babelgene" : Faster but fewer species and genes. Also gives consensus scores for each gene mapping based on a several different data sources. no_cores Number of cores to parallelise bootstrapping reps over. geneSizeControl Whether you want to control for GC content and transcript length. Recommended if the gene list originates from genetic studies (Default: FALSE). If set to TRUE, then hits must be from humans. controlledCT [Optional] If not NULL, and instead is the name of a cell type, then the bootstrapping controls for expression within that cell type. mtc_method Multiple-testing correction method (passed to p.adjust). sort_results Sort enrichment results from smallest to largest p-values. standardise_sct_data Should sct_data be standardised? if TRUE: When sctSpecies!=output_species the sct_data will be checked for object formatting and the genes will be converted to the orthologs of the output_species with standardise_ctd (which calls map_genes internally). When sctSpecies==output_species, the sct_data will be checked for object formatting with standardise_ctd, but the gene names will remain untouched. standardise_hits Should hits be standardised? If TRUE: When genelistSpecies!=output_species, the genes will be converted to the orthologs of the output_species with convert_orthologs. When genelistSpecies==output_species, the genes will be standardised with map_genes. If FALSE, hits will be passed on to subsequent steps as-is. localHub If working offline, add argument localHub=TRUE to work with a local, non-updated hub; It will only have resources available that have previously been downloaded. If offline, Please also see BiocManager vignette section on offline use to ensure proper functionality. store_gene_data Store sampled gene data for every bootstrap iteration. When the number of bootstrap reps is very high (>=100k) and/or the number of genes in hits is very high, you may want to set store_gene_data=FALSE to avoid using excessive amounts of CPU memory.

Details

The gene_data should be a data frame that contains a column of gene list names (e.g. the column may be called "hpo_name"), and a column of genes (e.g. "gene_symbol"). For example:

hpo_name	gene_symbol
"Abnormal heart"	gene X
"Abnormal heart"	gene Y
"Poor vision"	gene Z
"Poor vision"	gene Y
etc...

For more information on this see docs for get_gene_list (get_gene_lists).

Value

All results as a dataframe.

Examples

gene_data <- HPOExplorer::load_phenotype_to_genes()
list_names <- unique(gene_data$hpo_id)[seq(5)]
ctd <- load_example_ctd()
all_results <- gen_results(ctd = ctd,
                           gene_data = gene_data,
                           list_names = list_names,
                           reps = 10)

neurogenomics/MultiEWCE documentation built on April 22, 2024, 6:22 a.m.