# Rules for structuring isolate resources and filling in their properties

@prefix vg: <http://www.southdown.it/abk/onto/viral-genomics#>.

# rule1: Ensures a virus isolate resource exists for each sequence record.
# The virus isolate is created if it does not exist.
# The virus isolate is linked to linked to all its sequence record by the
# property vg:hasSequenceRecord
#
# To determine the isolate URI, the isolate name is normalized, which means
# that it is made lowercase, and all whitespace is removed. Any special
# characters other than '/' are also removed. Hence, isolate names containing
# "Hong Kong" "HongKong", Hongkong" and "Hong-Kong" will give same id. The
# idea is to overcome ids with alternative spellings.

[rule1: (?rec rdf:type   vg:SequenceRecord)
        (?rec vg:isolate ?isolateId)
        normalizeIsolate(?isolateId, ?nIsolateId)
        uriConcat('urn:abk:isolate:', ?nIsolateId, ?isolateUri)
        -> 
        (?isolateUri rdf:type             vg:VirusIsolate)
        (?isolateUri vg:hasSequenceRecord ?rec)
]


# rule2: Copy properties for isolate name, subtype, year, country and host
# from sequence record to virus isolate. This may copy multiple values from 
# different sequence records to the same virus isolate. The idea is that 
# these multiple values will be detected later, as the properties are 
# functional.

[rule2: (?isolateUri vg:hasSequenceRecord ?rec)
        (?rec ?prop ?value)
        oneOf(?prop, vg:isolate, vg:virusSubtype, vg:year, vg:country, vg:hostOrganism)
        -> 
        (?isolateUri ?prop ?value)
]