#! /usr/bin/env guile
!#

(use-modules (srfi srfi-1)
             (srfi srfi-26)
             (ice-9 match)
             (ice-9 regex)
             (dump strings)
             (dump sql)
             (dump triples)
             (dump special-forms))



(define %connection-settings
  (call-with-input-file (list-ref (command-line) 1)
    read))


(define (remap-species-identifiers str)
  "This procedure remaps identifiers to standard binominal. Obviously this should
   be sorted by correcting the database!"
  (match str
    ["Fly (Drosophila melanogaster dm6)" "Drosophila melanogaster"]
    ["Oryzias latipes (Japanese medaka)" "Oryzias latipes"]
    ["Macaca mulatta" "Macaca nemestrina"]
    ["Bat (Glossophaga soricina)" "Glossophaga soricina"]
    [str str]))


(define-dump dump-gene-chip
  (tables (GeneChip
           (left-join Species "USING (SpeciesId)")))
  (schema-triples
   (gnc:geneChip a skos:Concept)
   (gnc:geneChip
    skos:description
    "This is a set of controlled terms that are used to describe a given gene chip/platform")
   (gnt:hasGeoSeriesId rdfs:domain gnc:platform)
   (gnt:belongsToSpecies a owl:ObjectProperty)
   (gnt:belongsToSpecies skos:definition "This resource belongs to this given species")
   (gnt:belongsToSpecies rdfs:domain gnc:geneChip)
   (gnt:hasGeoSeriesId rdfs:domain gnc:geneChip)
   (gnt:hasGOTreeValue a owl:ObjectProperty)
   (gnt:hasGOTreeValue skos:definition "This resource the following GO tree value")
   (gnt:hasGOTreeValue rdfs:domain gnc:geneChip))
  (triples (string->identifier "platform" (field GeneChip Name))
    (set rdf:type 'gnc:geneChip)
    (set rdfs:label (field GeneChip GeneChipName))
    (set skos:prefLabel (field GeneChip Name))
    (set skos:altLabel (field ("IF(GeneChip.GeneChipName != GeneChip.Title, Title, NULL)"
                               Title)))
    (set gnt:hasGOTreeValue (field GeneChip Go_tree_value))
    (set gnt:belongsToSpecies
         (string->identifier "" (remap-species-identifiers (field Species Fullname))
                             #:separator ""
                             #:proc string-capitalize-first))
    (set gnt:hasGeoSeriesId
         (ontology 'geoSeries:
                   (string-trim-both (field GeneChip GeoPlatform))))))

;; Molecular Traits are also referred to as ProbeSets
(define-dump dump-probesetfreeze
  (tables (ProbeSetFreeze
           (left-join InfoFiles "ON InfoFiles.InfoPageName = ProbeSetFreeze.Name")
           (left-join ProbeFreeze "USING (ProbeFreezeId)")
           (left-join AvgMethod "ON AvgMethod.AvgMethodId = ProbeSetFreeze.AvgID")
           (left-join InbredSet "ON ProbeFreeze.InbredSetId = InbredSet.Id")
           (left-join Tissue "ON ProbeFreeze.TissueId = Tissue.TissueId"))
          "WHERE ProbeSetFreeze.public > 0 AND InfoFiles.InfoPageName IS NULL GROUP BY ProbeFreeze.Id")
  (schema-triples
   (gnc:probeset rdfs:subClassOf gnc:dataset)
   (gnt:usesNormalization rdfs:domain gnc:probeset)
   (gnt:usesDataScale rdfs:domain gnc:probeset)
   (gnt:usesDataScale a owl:ObjectProperty)
   (gnt:usesDataScale skos:definition "Thi data scale this resource uses"))
  (triples
      (string->identifier
       ""
       (regexp-substitute/global
        #f "[^A-Za-z0-9:]"
        (field ProbeSetFreeze Name)
        'pre "_" 'post)
       #:separator ""
       #:proc string-capitalize-first)
    (set rdf:type 'gnc:probeset)
    (set gnt:usesNormalization
         (string->identifier "avgmethod"
                             ;; If AvgMethodName is NULL, assume N/A.
                             (if (string-blank? (field AvgMethod Name AvgMethodName))
                                 "N/A" (field AvgMethod Name AvgMethodName))))
    (set dct:title (field ProbeSetFreeze FullName))
    (set rdfs:label (field ProbeSetFreeze Name))
    (set skos:altLabel (field ProbeSetFreeze Name2))
    (set skos:prefLabel (field ProbeSetFreeze ShortName))
    (set dct:created (annotate-field
                      (field ProbeSetFreeze CreateTime)
                      '^^xsd:datetime))
    (set gnt:usesDataScale (field ProbeSetFreeze DataScale))
    (set gnt:hasTissue
         (string->identifier
          "tissue"
          (field Tissue Short_Name)))
    (set gnt:belongsToInbredSet
         (string->identifier "inbredSet" (field InbredSet Name InbredSetName)))))



(dump-with-documentation
 (name "Probeset freeze metadata")
 (connection %connection-settings)
 (table-metadata? #f)
 (prefixes
  '(("geoSeries:" "<http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=>")
    ("gn:" "<http://genenetwork.org/id/>")
    ("gnc:" "<http://genenetwork.org/category/>")
    ("dct:" "<http://purl.org/dc/terms/>")
    ("owl:" "<http://www.w3.org/2002/07/owl#>")
    ("skos:" "<http://www.w3.org/2004/02/skos/core#>")
    ("gnt:" "<http://genenetwork.org/term/>")
    ("rdf:" "<http://www.w3.org/1999/02/22-rdf-syntax-ns#>")
    ("rdfs:" "<http://www.w3.org/2000/01/rdf-schema#>")
    ("xsd:" "<http://www.w3.org/2001/XMLSchema#>")))
 (inputs
  (list dump-gene-chip
        dump-probesetfreeze))
 (outputs
  '(#:documentation "./docs/dump-gene-chip.md"
    #:rdf "./verified-data/dump-probesetfreeze.ttl")))