Extract functional terms enriched in the DE genes, based on topGO

A wrapper for extracting functional GO terms enriched in the DE genes, based on the algorithm and the implementation in the topGO package

Usage

run_topGO(
  de_container = NULL,
  res_de = NULL,
  de_genes = NULL,
  bg_genes = NULL,
  top_de = NULL,
  FDR_threshold = 0.05,
  min_counts = 0,
  ontology = "BP",
  annot = annFUN.org,
  mapping = "org.Mm.eg.db",
  gene_id = "symbol",
  full_names_in_rows = TRUE,
  add_gene_to_terms = TRUE,
  de_type = "up_and_down",
  topGO_method2 = "elim",
  do_padj = FALSE,
  verbose = TRUE
)

Arguments

de_container

An object containing the data for a Differential Expression workflow (e.g. DESeq2, edgeR or limma). Currently, this can be a DESeqDataSet object, normally obtained after running your data through the DESeq2 framework.

res_de

An object containing the results of the Differential Expression analysis workflow (e.g. DESeq2, edgeR or limma). Currently, this can be a DESeqResults object created using the DESeq2 framework.

de_genes

A vector of (differentially expressed) genes

bg_genes

A vector of background genes, e.g. all (expressed) genes in the assays

top_de

numeric, how many of the top differentially expressed genes to use for the enrichment analysis. Attempts to reduce redundancy. Assumes the data is sorted by padj (default in DESeq2).

FDR_threshold

The pvalue threshold to us for counting genes as de. Default is 0.05

min_counts

numeric, min number of counts a gene needs to have to be included in the geneset that the de genes are compared to. Default is 0, recommended only for advanced users.

ontology

Which Gene Ontology domain to analyze: BP (Biological Process), MF (Molecular Function), or CC (Cellular Component)

annot

Which function to use for annotating genes to GO terms. Defaults to annFUN.org

mapping

Which org.XX.eg.db package to use for annotation - select according to the species

gene_id

Which format the genes are provided. Defaults to symbol, could also be entrez or ENSEMBL

full_names_in_rows

Logical, whether to display or not the full names for the GO terms

add_gene_to_terms

Logical, whether to add a column with all genes annotated to each GO term

de_type

One of: 'up', 'down', or 'up_and_down' Which genes to use for GOterm calculations: upregulated, downregulated or both

topGO_method2

Character, specifying which of the methods implemented by topGO should be used, in addition to the classic algorithm. Defaults to elim.

do_padj

Logical, whether to perform the adjustment on the p-values from the specific topGO method, based on the FDR correction. Defaults to FALSE, since the assumption of independent hypotheses is somewhat violated by the intrinsic DAG-structure of the Gene Ontology Terms

verbose

Logical, whether to add messages telling the user which steps were taken

Value

A table containing the computed GO Terms and related enrichment scores

Details

Allowed values assumed by the topGO_method2 parameter are one of the following: elim, weight, weight01, lea, parentchild. For more details on this, please refer to the original documentation of the topGO package itself

See also

topGO::topGOdata-class() and topGO::runTest() for the class objects and underlying methods

Other Enrichment functions: run_cluPro(), run_goseq()

Examples

library("macrophage")
library("DESeq2")
data(gse, package = "macrophage")

dds_macrophage <- DESeqDataSet(gse, design = ~ line + condition)
#> using counts and average transcript lengths from tximeta
rownames(dds_macrophage) <- substr(rownames(dds_macrophage), 1, 15)
keep <- rowSums(counts(dds_macrophage) >= 10) >= 6
dds_macrophage <- dds_macrophage[keep, ]
dds_macrophage <- DESeq(dds_macrophage)
#> estimating size factors
#> using 'avgTxLength' from assays(dds), correcting for library size
#> estimating dispersions
#> gene-wise dispersion estimates
#> mean-dispersion relationship
#> final dispersion estimates
#> fitting model and testing

data(res_de_macrophage, package = "mosdef")

library("AnnotationDbi")
library("org.Hs.eg.db")
library("topGO")
#> Loading required package: graph
#> Loading required package: GO.db
#> Loading required package: SparseM
#> 
#> groupGOTerms: 	GOBPTerm, GOMFTerm, GOCCTerm environments built.
#> 
#> Attaching package: ‘topGO’
#> The following object is masked from ‘package:GenomicFeatures’:
#> 
#>     genes
#> The following object is masked from ‘package:IRanges’:
#> 
#>     members
topgoDE_macrophage <- run_topGO(
  de_container = dds_macrophage,
  res_de = res_macrophage_IFNg_vs_naive,
  ontology = "BP",
  mapping = "org.Hs.eg.db",
  gene_id = "symbol",
)
#> 'select()' returned 1:many mapping between keys and columns
#> 'select()' returned 1:many mapping between keys and columns
#> Your dataset has 1024 DE genes.
#> You selected 1024 (100.00%) genes for the enrichment analysis.
#> You are analyzing up_and_down-regulated genes in the `res_de` container
#> Warning: NAs introduced by coercion
#> 6071 GO terms were analyzed. Not all of them are significantly enriched.
#> We suggest further subsetting the output list by for example: 
#> using a pvalue cutoff in the column: 
#> 'p.value_elim'.