Ontology tools for computational genomic data science

ontoProc

This package enables users to explore and visualize relationships among concepts enumerated in biomedical ontologies.

Basic representations

ontoProc uses AnnotationHub to manage ontology representations. We can retrieve a recent version of the UBERON ontology as follows:

library(AnnotationHub)
ah = AnnotationHub()
query(ah, "ontoProc")

## AnnotationHub with 29 records
## # snapshotDate(): 2022-12-16
## # $dataprovider: NA
## # $species: NA
## # $rdataclass: Rda
## # additional mcols(): taxonomyid, genome, description,
## #   coordinate_1_based, maintainer, rdatadateadded, preparerclass, tags,
## #   rdatapath, sourceurl, sourcetype 
## # retrieve records with, e.g., 'object[["AH97934"]]' 
## 
##              title              
##   AH97934  | caro               
##   AH97935  | cellLineOnto       
##   AH97936  | cellOnto           
##   AH97937  | cellosaurusOnto    
##   AH97938  | chebi_full         
##   ...        ...                
##   AH107277 | hcaoOnto_2022.07.19
##   AH107278 | mondo_2022.08.01   
##   AH107279 | patoOnto_2022.08.10
##   AH107280 | PROnto_67          
##   AH107281 | uberon_2022.06.30

library(ontoProc)
ub = getOnto("uberon")
head(names(ub))

## [1] "id"        "name"      "parents"   "children"  "ancestors" "obsolete"

ub$name[1:10]

##                         BFO:0000001                         BFO:0000002 
##                            "entity"                        "continuant" 
##                         BFO:0000003                         BFO:0000004 
##                         "occurrent"            "independent continuant" 
##                         BFO:0000006                         BFO:0000015 
##                    "spatial region"                           "process" 
##                         BFO:0000016                         BFO:0000017 
##                       "disposition"                 "realizable entity" 
##                         BFO:0000019                         BFO:0000020 
##                           "quality" "specifically dependent continuant"

Visualization

We take advantage of Daniel Greene’s ontologyIndex suite of packages to produce visualizations of concept relationships.

cl = ontoProc::getOnto('cellOnto')
cl3k = c("CL:0000492", "CL:0001054", "CL:0000236", "CL:0000625",
    "CL:0000576", "CL:0000623", "CL:0000451", "CL:0000556")
ontoProc::onto_plot2(cl, cl3k)

Interactive app

The ctmarks function defines a shiny app that addresses cell-type-defining marker genes. Users select a cell type, and its position in the Cell Ontology network is rendered.

Using available inter-ontology relationships, protein and transcriptional annotations for the cell type are tabulated.

pogos

The pogos package confronts the relationships between ontologies for cells and cell lines and compounds used in pharmacology and pharmacogenomics.

The Cancer Cell Line Encyclopedia (CCLE) includes datasets describing sensitivity of cell lines to drugs. pogos provides tools for querying the PharmacoDB system and rendering sensitivity results from CCLE.

library(pogos)
iri = iriCCLE()
iri

## DRTraceSet for 5 cell lines, drug Irinotecan, dataset CCLE

plot(iri)

Other tools

Bioconductor includes the rols package that interrogates the EBI Ontology Lookup Service.