Analyte? Target Entity?

turbomam commented 4 years ago

The TURBO assay template columns are practically identical to the OBI assay template columns.

As of today, the TURBO Assay template is constructed so that the glucose in the blood etc. appears in the target entity column.
ideally, anything that could bear an analyte role would appear in the existing analyte column.
there has been some talk within OBI about adding a measurand column to the OBI template. That would be a good place for things like neutrophils in a blood evaluant.

I am working on a SPARQL query that will migrate or add analute eligible entities (molecular entities or atoms) into the analyte column

It might also be possible to make that distinction within the script that builds the template.

In either case, handling target entities like 'oxycodone or oxymorphone' will be a challenge

turbomam commented 4 years ago

ROBOT column syntax:

target entity: C has_specified_output some ('is about' some %)
analyte: C (has_specified_input some %) and (realizes some ('analyte role' and ('inheres in' some %)))

turbomam commented 4 years ago

Issues that affect analyte conversion (among other aspects of the programmatic template generation)

Can't identify analyte assays based on subClassOf assertion about target entity
Would Hermit even assert analyte role if an equivalent class definition was used?

Some non-reasoning solutions and lingering concerns:

https://github.com/PennTURBO/turbo-assay-templating/blob/master/pipeline/data/loinc_to_obo_mapping_reviewed.csv expresses mapping with labels (not IRIs)
Could always add IRIs to loinc_to_obo_mapping_reviewed.csv and process with https://github.com/PennTURBO/turbo-assay-templating/blob/master/pipeline/turbo_assay_template_generation.R
Mappings containing "complex" class expressions (oxycodone and oxymorphone) would require parsing

Example requiring conversion to analyte: Quantitiative assay for '(-)-cotinine' in 'urine specimen' ['scalar measurement datum' and ('has measurement unit label' some 'mass per unit volume')]

SPARQL approach

PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
insert {
    graph <http://example.com/resource/analyte> {
        ?assay rdfs:subClassOf  [ owl:intersectionOf [  rdf:first  [ a owl:Restriction ;
                                                                     owl:onProperty <http://purl.obolibrary.org/obo/BFO_0000055> ;
                                                                     owl:someValuesFrom [ owl:intersectionOf  [ rdf:first  <http://purl.obolibrary.org/obo/OBI_0000275> ]] ] ;
                                                                                                                rdf:rest  [ rdf:first [ a owl:Restriction ;
                                                                                                                                        owl:onProperty <http://purl.obolibrary.org/obo/RO_0000081> ;
                                                                                                                                        owl:someValuesFrom  [ owl:intersectionOf  [ rdf:first ?targent ] ] ] ]  ]  ] .
        ?assay rdfs:subClassOf  [ owl:intersectionOf [  rdf:first  [ a owl:Restriction ;
                                                                     owl:onProperty <http://purl.obolibrary.org/obo/OBI_0000293> ;
                                                                     owl:someValuesFrom [ owl:intersectionOf  [ rdf:first  ?targent ]] ] ;
                                                                                                                rdf:rest  [ rdf:first [ a owl:Restriction ;
                                                                                                                                        owl:onProperty <http://purl.obolibrary.org/obo/RO_0000087> ;
                                                                                                                                        owl:someValuesFrom  [ owl:intersectionOf  [ rdf:first <http://purl.obolibrary.org/obo/OBI_0000275> ] ] ] ]  ]  ] .
        <http://purl.obolibrary.org/obo/OBI_0000275> rdfs:label "analyte role" ;
                                                                               rdfs:subClassOf <http://purl.obolibrary.org/obo/OBI_0002444> .
        <http://purl.obolibrary.org/obo/OBI_0002444> rdfs:label "measurand role" ;
                                                                                 rdfs:subClassOf <http://purl.obolibrary.org/obo/BFO_0000023> .
    }
} where {
    ?assay rdfs:subClassOf  [ a owl:Restriction ;
                              owl:onProperty <http://purl.obolibrary.org/obo/OBI_0000299> ;
                              owl:someValuesFrom [ a owl:Restriction ;
                                                   owl:onProperty <http://purl.obolibrary.org/obo/IAO_0000136> ;
                                                   owl:someValuesFrom ?targent ;
                                                   ]  ] .
    values ?eligible {
        <http://purl.obolibrary.org/obo/CHEBI_23367> <http://purl.obolibrary.org/obo/CHEBI_33250>
    }
    ?targent rdfs:subClassOf* <http://purl.obolibrary.org/obo/CHEBI_23367> .
}

Added 12424 statements. Update took 3.8s, moments ago.

WITH NO ELIGIBILITY CONSTRAINT: Added 12814 statements. Update took 0.3s, moments ago.

The OBI assays template "and"s the "has analyte input" and "realizes analyte role axioms" together.Here, they are asserted independently. Is that acceptable?

Positive control

Quantitiative assay for '(-)-cotinine' in 'urine specimen' ['scalar measurement datum' and ('has measurement unit label' some 'mass per unit volume')]

Not currently updated

Some are good negative controls, but some are missed by the current algorithm for reasons discussed above or...

PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
PREFIX owl: <http://www.w3.org/2002/07/owl#>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
select * where {
    ?assay rdfs:subClassOf* <http://purl.obolibrary.org/obo/OBI_0000070> ;
                          rdfs:label ?l .
    minus {
        ?assay rdfs:subClassOf  [ owl:intersectionOf [  rdf:first  [ a owl:Restriction ;
                                                                     owl:onProperty <http://purl.obolibrary.org/obo/BFO_0000055> ;
                                                                     owl:someValuesFrom [ owl:intersectionOf  [ rdf:first  <http://purl.obolibrary.org/obo/OBI_0000275> ]] ] ;
                                                                                                                rdf:rest  [ rdf:first [ a owl:Restriction ;
                                                                                                                                        owl:onProperty <http://purl.obolibrary.org/obo/RO_0000081> ;
                                                                                                                                        owl:someValuesFrom  [ owl:intersectionOf  [ rdf:first ?targent ] ] ] ]  ]  ] .
        ?assay rdfs:subClassOf  [ owl:intersectionOf [  rdf:first  [ a owl:Restriction ;
                                                                     owl:onProperty <http://purl.obolibrary.org/obo/OBI_0000293> ;
                                                                     owl:someValuesFrom [ owl:intersectionOf  [ rdf:first  ?targent ]] ] ;
                                                                                                                rdf:rest  [ rdf:first [ a owl:Restriction ;
                                                                                                                                        owl:onProperty <http://purl.obolibrary.org/obo/RO_0000087> ;
                                                                                                                                        owl:someValuesFrom  [ owl:intersectionOf  [ rdf:first <http://purl.obolibrary.org/obo/OBI_0000275> ] ] ] ]  ]  ] .
    }
}

l	targent
http://purl.obolibrary.org/obo/OBI_0000070	"assay"@en
turbo:TURBO_3000001	Quantitiative assay for 'immunoglobulin complex' and 'has disposition to bind' some 'acetylcholine receptor binding' in 'blood serum specimen' ['scalar measurement datum' and ('has measurement unit label' some 'unit of molarity')]
turbo:TURBO_3000002	Quantitiative assay for 'IgG immunoglobulin complex' and 'has disposition to bind' some 'Polistes' in 'blood serum specimen' ['scalar measurement datum' and ('has measurement unit label' some 'mass per unit volume')]
turbo:TURBO_3000003	Quantitiative assay for 'IgG immunoglobulin complex' and 'has disposition to bind' some 'Dolichovespula arenaria' in 'blood serum specimen' ['scalar measurement datum' and ('has measurement unit label' some 'mass per unit volume')]
turbo:TURBO_3000004	Quantitiative assay for 'IgG immunoglobulin complex' and 'has disposition to bind' some 'Dolichovespula maculata' in 'blood serum specimen' ['scalar measurement datum' and ('has measurement unit label' some 'mass per unit volume')]
turbo:TURBO_3000174	Quantitiative assay for 'osmolality' in ('blood serum specimen' or 'blood plasma specimen') ['scalar measurement datum' and ('is about' some 'osmolality')]
turbo:TURBO_3000175	Quantitiative assay for 'osmolality' in 'feces specimen' ['scalar measurement datum' and ('is about' some 'osmolality')]
turbo:TURBO_3000176	Quantitiative assay for 'osmolality' in 'urine specimen' ['scalar measurement datum' and ('is about' some 'osmolality')]
turbo:TURBO_3000207	Quantitiative assay for 'growth hormone' in ('blood serum specimen' or 'blood plasma specimen') ['scalar measurement datum' and ('has measurement unit label' some 'mass per unit volume')]
turbo:TURBO_3000258	Quantitiative assay for ('amitriptyline' and 'nortriptyline') in ('blood serum specimen' or 'blood plasma specimen') ['scalar measurement datum' and ('has measurement unit label' some 'mass per unit volume')]
turbo:TURBO_3000309	Quantitiative assay for 'propoxyphene' in 'urine specimen' ['scalar measurement datum' and ('has measurement unit label' some 'mass per unit volume')]
turbo:TURBO_3000322	Quantitiative assay for ('doxepin' and 'desmethyldoxepin') in ('blood serum specimen' or 'blood plasma specimen') ['scalar measurement datum' and ('has measurement unit label' some 'mass per unit volume')]
turbo:TURBO_3000327	Quantitiative assay for 'fluoxetine' in ('blood serum specimen' or 'blood plasma specimen') ['scalar measurement datum' and ('has measurement unit label' some 'mass per unit volume')]
turbo:TURBO_3000345	Quantitiative assay for 'nicotine' in ('blood serum specimen' or 'blood plasma specimen') ['scalar measurement datum' and ('has measurement unit label' some 'mass per unit volume')]
turbo:TURBO_3000376	Quantitiative assay for ('calcidiol' and 'vitamin D2') in ('blood serum specimen' or 'blood plasma specimen') ['scalar measurement datum' and ('has measurement unit label' some 'mass per unit volume')]
turbo:TURBO_3000389	Quantitiative assay for ('amphetamine' and 'methamphetamine') in 'urine specimen' ['scalar measurement datum' and ('has measurement unit label' some 'mass per unit volume')]
turbo:TURBO_3000423	Quantitiative assay for 'adalimumab' in ('blood serum specimen' or 'blood plasma specimen') ['scalar measurement datum' and ('has measurement unit label' some 'mass per unit volume')]
turbo:TURBO_3000428	Quantitiative assay for ('imipramine' and 'desipramine') in ('blood serum specimen' or 'blood plasma specimen') ['scalar measurement datum' and ('has measurement unit label' some 'mass per unit volume')]

turbomam commented 4 years ago

Queries above have been updated. See pipeline/config/*.rq

PennTURBO / turbo-assay-templating