Ancestor (generalization) retrieval

go_ancestors utilizes the pre-computed ‘entailed_edge’ table in the Semantic SQL representation of GO.

go_ancestors("GO:0006954", relations = GO_RELATIONS["is_a"]) |>
  inner_join(
    go_terms() |> dplyr::select(ancestor_id = id, label),
    by = "ancestor_id"
  )

## # A query:  ?? x 4
## # Database: DuckDB 1.5.6 [unknown@Linux 7.0.0-1012-azure:R 4.6.1/:memory:]
##   id         ancestor_id relation        label               
##   <chr>      <chr>       <chr>           <chr>               
## 1 GO:0006954 GO:0006950  rdfs:subClassOf response to stress  
## 2 GO:0006954 GO:0006952  rdfs:subClassOf defense response    
## 3 GO:0006954 GO:0008150  rdfs:subClassOf biological_process  
## 4 GO:0006954 GO:0050896  rdfs:subClassOf response to stimulus

Descendant

go_descendants("GO:0006954", relations = GO_RELATIONS["is_a"]) |>
  inner_join(
    go_terms() |> dplyr::select(descendant_id = id, label),
    by = "descendant_id"
  )

## # A query:  ?? x 4
## # Database: DuckDB 1.5.6 [unknown@Linux 7.0.0-1012-azure:R 4.6.1/:memory:]
##    id         descendant_id relation        label                               
##    <chr>      <chr>         <chr>           <chr>                               
##  1 GO:0006954 GO:0001660    rdfs:subClassOf fever generation                    
##  2 GO:0006954 GO:0001788    rdfs:subClassOf antibody-dependent cellular cytotox…
##  3 GO:0006954 GO:0001794    rdfs:subClassOf type IIa hypersensitivity           
##  4 GO:0006954 GO:0001795    rdfs:subClassOf type IIb hypersensitivity           
##  5 GO:0006954 GO:0001802    rdfs:subClassOf type III hypersensitivity           
##  6 GO:0006954 GO:0001806    rdfs:subClassOf type IV hypersensitivity            
##  7 GO:0006954 GO:0002254    rdfs:subClassOf kinin cascade                       
##  8 GO:0006954 GO:0002255    rdfs:subClassOf tissue kallikrein-kinin cascade     
##  9 GO:0006954 GO:0002353    rdfs:subClassOf plasma kallikrein-kinin cascade     
## 10 GO:0006954 GO:1902483    rdfs:subClassOf cytotoxic T cell pyroptotic cell de…
## # ℹ more rows

Quick overview of RDF and ontological entailments

Briefly, the fundamental representation of Gene Ontology is RDF (Resource Description Format). Each ontological assertion (statement) has the form of an ordered triple (subject, predicate, object).

All the implications of statements in Gene Ontology are available in the Semantic SQL representation. “Implication” here refers to derivability in a logical system implemented in an OWL reasoner. See OAK (ontology access kit) for more background.

Common predicates

To express certain concepts precisely, references to external ontologies are employed.

go_entailed_edges() |> dplyr::select(predicate) |> 
   group_by(predicate) |> summarise(n=n()) |>
   arrange(desc(n))

## # A query:    ?? x 2
## # Database:   DuckDB 1.5.6 [unknown@Linux 7.0.0-1012-azure:R 4.6.1/:memory:]
## # Ordered by: desc(n)
##    predicate             n
##    <chr>             <dbl>
##  1 RO:0002131      1378323
##  2 rdfs:subClassOf 1367716
##  3 BFO:0000051     1053425
##  4 RO:0000057       659418
##  5 RO:0018038       481870
##  6 BFO:0000050      350002
##  7 RO:0002162       345293
##  8 RO:0004007       309651
##  9 RO:0018033       308277
## 10 RO:0018034       228886
## # ℹ more rows

Let’s check on the interpretation of a few of these frequently encountered CURIEs.

roref = semsql_connect(ontology="ro")

## Connected to SemanticSQL database: /github/home/.cache/R/BiocFileCache/15907c2183ce_ro.db

## Primary ontology prefix: RO

goref = semsql_connect(ontology="go")

## Connected to SemanticSQL database: /github/home/.cache/R/BiocFileCache/159033af6a15_go.db

## Primary ontology prefix: GO

tbl(roref@con, "rdfs_label_statement") |> filter(subject %in% c("RO:0002131", "RO:0000057",
   "RO:0018038")) |> as.data.frame() |> DT::datatable()

tbl(goref@con, "rdfs_label_statement") |> filter(subject %in% c("BFO:0000050", "BFO:0000051")) |>
  as.data.frame() |> DT::datatable()

Decoding external predicate CURIEs

We can use ontoProc2 to help resolve details of predicates used to state relationships beyond those related to class membership (“is a”, “part of”).

library(ontoProc2)
library(DT)
bspo_ref = semsql_connect(ontology="bspo")
bspo_ref

## <SemsqlConn>  prefix: BSPO  | labeled terms: 402

tbl(bspo_ref@con, "statements") |> head(20) |>
   collect() |> datatable()

library(ontoProc2)
library(DT)
bspo_ref = semsql_connect(ontology="bspo")
bspo_ref

## <SemsqlConn>  prefix: BSPO  | labeled terms: 402

tbl(bspo_ref@con, "statements") |> filter(subject=="BSPO:0000098")  |> 
   collect() |> datatable()

AnnotationDbi “select”

The select procedure of AnnotationDbi can be replaced by a filter on go_terms.

library(GO.db)

## Loading required package: AnnotationDbi

## Loading required package: stats4

## Loading required package: BiocGenerics

## Loading required package: generics

## 
## Attaching package: 'generics'

## The following object is masked from 'package:dplyr':
## 
##     explain

## The following objects are masked from 'package:base':
## 
##     as.difftime, as.factor, as.ordered, intersect, is.element, setdiff,
##     setequal, union

## 
## Attaching package: 'BiocGenerics'

## The following object is masked from 'package:dplyr':
## 
##     combine

## The following objects are masked from 'package:stats':
## 
##     IQR, mad, sd, var, xtabs

## The following object is masked from 'package:utils':
## 
##     data

## The following objects are masked from 'package:base':
## 
##     anyDuplicated, aperm, append, as.data.frame, basename, cbind,
##     colnames, dirname, do.call, duplicated, eval, evalq, Filter, Find,
##     get, grep, grepl, is.unsorted, lapply, Map, mapply, match, mget,
##     order, paste, pmax, pmax.int, pmin, pmin.int, Position, rank,
##     rbind, Reduce, rownames, sapply, saveRDS, scale, sequence, table,
##     tapply, transform, unique, unsplit, which.max, which.min

## Loading required package: Biobase

## Welcome to Bioconductor
## 
##     Vignettes contain introductory material; view with
##     'browseVignettes()'. To cite Bioconductor, see
##     'citation("Biobase")', and for packages 'citation("pkgname")'.

## Loading required package: IRanges

## Loading required package: S4Vectors

## 
## Attaching package: 'S4Vectors'

## The following objects are masked from 'package:dplyr':
## 
##     first, rename

## The following object is masked from 'package:utils':
## 
##     findMatches

## The following objects are masked from 'package:base':
## 
##     expand.grid, I, unname

## 
## Attaching package: 'IRanges'

## The following objects are masked from 'package:dplyr':
## 
##     collapse, desc, slice

## 
## Attaching package: 'AnnotationDbi'

## The following object is masked from 'package:dplyr':
## 
##     select

## 

AnnotationDbi::select(GO.db, keys="GO:0006954", 
  columns=c("TERM", "DEFINITION", "ONTOLOGY"))

## 'select()' returned 1:1 mapping between keys and columns

##         GOID                  TERM
## 1 GO:0006954 inflammatory response
##                                                                                                                                                                                                                                                                           DEFINITION
## 1 The immediate defensive reaction (by vertebrate tissue) to infection or injury caused by chemical or physical agents. The process is characterized by local vasodilation, extravasation of plasma into intercellular spaces and accumulation of white blood cells and macrophages.
##   ONTOLOGY
## 1       BP

Similar content is provided using go_terms:

GO.ddb::go_terms() |> filter(id == "GO:0006954") |> as.data.frame()

##           id                 label           ontology
## 1 GO:0006954 inflammatory response biological_process
##                                                                                                                                                                                                                                                                           definition
## 1 The immediate defensive reaction (by vertebrate tissue) to infection or injury caused by chemical or physical agents. The process is characterized by local vasodilation, extravasation of plasma into intercellular spaces and accumulation of white blood cells and macrophages.
##   deprecated
## 1      FALSE

Environments

The GO.db package includes environments for efficient management of key-value pairs.

We’ll illustrate migration activities for one, the management of generalization of a selected concept via traversal of the GO graph. Here we use GO:0000112, denoting nucleotide-excision repair factor 3 complex. A view of the graph is provided here. Note that we do not include in this example any edges other than “is a” and “part of”, although other edge types can be explored with the “entailed_edge” table.

QuickGO view

ls("package:GO.db")

##  [1] "GO"            "GO_dbconn"     "GO_dbfile"     "GO_dbInfo"    
##  [5] "GO_dbschema"   "GO.db"         "GOBPANCESTOR"  "GOBPCHILDREN" 
##  [9] "GOBPOFFSPRING" "GOBPPARENTS"   "GOCCANCESTOR"  "GOCCCHILDREN" 
## [13] "GOCCOFFSPRING" "GOCCPARENTS"   "GOMAPCOUNTS"   "GOMFANCESTOR" 
## [17] "GOMFCHILDREN"  "GOMFOFFSPRING" "GOMFPARENTS"   "GOOBSOLETE"   
## [21] "GOSYNONYM"     "GOTERM"

get("GO:0000112", GOCCANCESTOR)

##  [1] "GO:0000109" "GO:0005575" "GO:0005622" "GO:0005634" "GO:0032991"
##  [6] "GO:0043226" "GO:0043227" "GO:0043229" "GO:0043231" "GO:0110165"
## [11] "GO:0140513" "all"

We use go_ancestors to obtain this information.

lk112 = go_ancestors("GO:0000112") |> as.data.frame()
lk112

##            id    ancestor_id        relation
## 1  GO:0000112    BFO:0000040 rdfs:subClassOf
## 2  GO:0000112    BFO:0000002 rdfs:subClassOf
## 3  GO:0000112    COB:0000080 rdfs:subClassOf
## 4  GO:0000112     GO:0032991 rdfs:subClassOf
## 5  GO:0000112     GO:0140513 rdfs:subClassOf
## 6  GO:0000112    BFO:0000004 rdfs:subClassOf
## 7  GO:0000112     GO:0005575 rdfs:subClassOf
## 8  GO:0000112     GO:0000109 rdfs:subClassOf
## 9  GO:0000112    BFO:0000002     BFO:0000050
## 10 GO:0000112    BFO:0000004     BFO:0000050
## 11 GO:0000112    BFO:0000040     BFO:0000050
## 12 GO:0000112 UBERON:0001062     BFO:0000050
## 13 GO:0000112     GO:0005575     BFO:0000050
## 14 GO:0000112 UBERON:0000465     BFO:0000050
## 15 GO:0000112     GO:0110165     BFO:0000050
## 16 GO:0000112 UBERON:0000061     BFO:0000050
## 17 GO:0000112     GO:0043226     BFO:0000050
## 18 GO:0000112     GO:0005622     BFO:0000050
## 19 GO:0000112     CL:0000000     BFO:0000050
## 20 GO:0000112     GO:0043227     BFO:0000050
## 21 GO:0000112     GO:0043229     BFO:0000050
## 22 GO:0000112     GO:0043231     BFO:0000050
## 23 GO:0000112     GO:0005634     BFO:0000050

Filtering down to the relevant GO terms:

lk112 |> dplyr::filter(grepl("^GO:", ancestor_id))

##            id ancestor_id        relation
## 1  GO:0000112  GO:0032991 rdfs:subClassOf
## 2  GO:0000112  GO:0140513 rdfs:subClassOf
## 3  GO:0000112  GO:0005575 rdfs:subClassOf
## 4  GO:0000112  GO:0000109 rdfs:subClassOf
## 5  GO:0000112  GO:0005575     BFO:0000050
## 6  GO:0000112  GO:0110165     BFO:0000050
## 7  GO:0000112  GO:0043226     BFO:0000050
## 8  GO:0000112  GO:0005622     BFO:0000050
## 9  GO:0000112  GO:0043227     BFO:0000050
## 10 GO:0000112  GO:0043229     BFO:0000050
## 11 GO:0000112  GO:0043231     BFO:0000050
## 12 GO:0000112  GO:0005634     BFO:0000050

In summary, migration of AnnotationDbi-dependent code should be possible, although different functions and data structures will be encountered. If a particular legacy computation seems infeasible with GO.ddb, please raise an issue. We do not plan to update AnnotationDbi or GO.db in Bioconductor 3.24 or beyond, although the GO.db package will remain available with 2026 content.

Term to descendant ids

Here’s a simple function to search for a regular expression in the ontology.

go_ids = function (regexpr, use = "label") 
{
    dplyr::filter(go_terms(), grepl(regexpr, !!sym(use)))
}

We use this to enumerate specialized concepts related to “gonad development”.

library(duckplyr)

## ✔ Overwriting dplyr methods with duckplyr methods.
## ℹ Turn off with `duckplyr::methods_restore()`.

gd = go_ids("gonad development") |>
  dplyr::select(id) |>
  collect() |>
  pull(id) |>
  go_descendants(include_self=TRUE) |>
  left_join(go_terms(), by = c("descendant_id" = "id")) |>
  dplyr::select(id = descendant_id, label)

## duckdb keeps downloaded extensions and secrets in a temporary directory:
## ℹ /tmp/Rtmp2mpXyA/duckdb
## This is removed when the R session ends.
## • Extensions are re-downloaded each session.
## • Secrets are lost.
## ℹ Run duckdb(shared_home = TRUE) (or create ~/.duckdb) to keep them (suitable for most users).
## ℹ Run duckdb(shared_home = FALSE) to accept the temporary directory (and silence this message).
## ℹ See ?duckdb_storage for details and alternatives.

gd

## # A query:  ?? x 2
## # Database: DuckDB 1.5.6 [unknown@Linux 7.0.0-1012-azure:R 4.6.1/:memory:]
##    id         label                                          
##    <chr>      <chr>                                          
##  1 GO:2000195 negative regulation of female gonad development
##  2 GO:2000872 positive regulation of progesterone secretion  
##  3 GO:2000196 positive regulation of female gonad development
##  4 GO:2000020 positive regulation of male gonad development  
##  5 GO:2000019 negative regulation of male gonad development  
##  6 GO:1905940 negative regulation of gonad development       
##  7 GO:2000870 regulation of progesterone secretion           
##  8 GO:2000194 regulation of female gonad development         
##  9 GO:1905939 regulation of gonad development                
## 10 GO:1905941 positive regulation of gonad development       
## # ℹ more rows

Term to human genes

This section depends on the availability of an experimental package. To get it, use BiocManager::install("vjcitn/GOAnno4Bioc").

Briefly get_gaf obtains a representation of the Gene Annotation Format resource for a model organism. We get a lazy representation of this resource, filter it to the GO terms derived in the previous section, and report on the distinct genes and aspects of the annotation assertions.

if (requireNamespace("GOAnno4Bioc")) {
  hu = GOAnno4Bioc::get_gaf("human")
  hulim = inner_join( hu, gd, by=c(go_id="id"), copy=TRUE, relationship = "many-to-many") |>
      dplyr::select(-db_reference,-evidence_code,-with_or_from,-date,-assigned_by) |> distinct()
  hulim
}

## Loading required namespace: GOAnno4Bioc