One-sided Fisher's exact test (over-representation) per GO term, with multiple-testing adjustment over the terms tested in the aspect. Aspects BP (biological process), MF (molecular function), and CC (cellular component) are recognised under all common naming conventions (`BP`/`P`/`biological_process`, etc.). Note: candidate genes from GWAS intervals are not independent (LD clusters, overlapping genes), so treat the p-values as exploratory.
Usage
hbr_go_enrichment(
genes_superior,
genes_all,
goslim_dt,
aspect = c("BP", "MF", "CC", "all"),
min_term_size = 5L,
min_superior = 2L,
p_adjust = "BH",
top_n = 50L,
restrict_to_annotated = FALSE,
verbose = TRUE
)Arguments
- genes_superior
Character vector of genes of interest.
- genes_all
Character vector of the full tested universe.
- goslim_dt
GOSlim table with columns `Gene_ID`, `GO_ID`, `GO_Aspect`, `GO_Term`.
- aspect
One of `"BP"`, `"MF"`, `"CC"`, `"all"`.
- min_term_size
Minimum term size in the universe.
- min_superior
Minimum superior carriers.
- p_adjust
Multiple-testing method.
- top_n
Return at most this many top-ranked terms.
- restrict_to_annotated
Restrict the universe to genes with at least one GO term in the aspect.
- verbose
Print progress.
Value
A `data.table` with columns `GO_ID`, `GO_Term`, `GO_Aspect`, `N_Superior`, `N_Universe`, `OR`, `P`, `Adjusted_P`.
Examples
goslim <- data.table::data.table(
Gene_ID = c("Os01g0100100", "Os01g0100200", "Os01g0100300"),
GO_ID = c("GO:0008150", "GO:0008150", "GO:0003674"),
GO_Aspect = c("P", "P", "F"),
GO_Term = c("biological_process", "biological_process",
"molecular_function"))
hbr_go_enrichment(
genes_superior = c("Os01g0100100", "Os01g0100200"),
genes_all = c("Os01g0100100", "Os01g0100200", "Os01g0100300"),
goslim_dt = goslim,
aspect = "BP",
min_term_size = 1L,
min_superior = 1L)
#> Enrichment: 2 superior / 3 universe genes | aspect = BP
#> 1 terms tested | 0 significant (FDR < 0.05)
#> GO_ID GO_Term GO_Aspect N_Superior N_Universe OR
#> <char> <char> <char> <int> <int> <num>
#> 1: GO:0008150 biological_process BP 2 2 Inf
#> P Adjusted_P
#> <num> <num>
#> 1: 0.3333333 0.3333333