#! /usr/bin/env guile
!#

(use-modules (srfi srfi-1)
             (srfi srfi-26)
             (ice-9 getopt-long)
             (ice-9 match)
             (ice-9 regex)
             (transform strings)
             (transform sql)
             (transform triples)
             (transform special-forms))

(define-transformer probeset
  (tables (ProbeSet
           (left-join GeneChip "ON GeneChip.Id = ProbeSet.ChipId")
           (left-join GeneList "ON GeneList.GeneID = ProbeSet.GeneId")
           (left-join GeneList_rn33 "ON GeneList.geneSymbol = ProbeSet.Symbol")
           (left-join Species "ON GeneChip.SpeciesId = Species.Id")))
  (schema-triples
   (gnc:pantherLink rdf:type gnc:ResourceLink)
   (gnc:pantherLink rdfs:label "PANTHER")
   (gnc:pantherLink rdfs:comments "Gene and protein data resources from Celera-ABI")
   (gnc:gnt:NCBIGeneLink rdfs:Class gnc:ResourceLink)
   (gnc:gnt:NCBIGeneLink rdfs:label "Gene")
   (gnc:gnt:NCBIGeneLink rdfs:comments "Info from NCBI Entrez Gene")
   (gnc:omimLink rdfs:Class gnc:ResourceLink)
   (gnc:omimLink rdfs:label "OMIM")
   (gnc:omimLink rdfs:comments "Summary from On Mendelion Inheritance in Man")
   (gnc:homologeneLink rdfs:Class gnc:ResourceLink)
   (gnc:homologeneLink rdfs:label "HomoloGene")
   (gnc:homologeneLink rdfs:comments "Find similar genes in other species")
   (gnc:uniprotLink rdfs:Class gnc:ResourceLink)
   (gnc:uniprotLink rdfs:label "UniProt")
   (gnc:uniprotLink rdfs:comments "UniProt")
   (gnc:stringLink rdfs:Class gnc:ResourceLink)
   (gnc:stringLink rdfs:label "STRING")
   (gnc:stringLink rdfs:comments "Protein interactions: known and inferred")
   (gnc:gtexLink rdfs:Class gnc:ResourceLink)
   (gnc:gtexLink rdfs:label "GTEx Portal")
   (gnc:gtexLink rdfs:comments "GTEx Portal")
   (gnc:ebiGwasLink rdfs:Class gnc:ResourceLink)
   (gnc:ebiGwasLink rdfs:label "EBI GWAS")
   (gnc:ebiGwasLink rdfs:comments "EBI GWAS")
   (gnc:genemaniaLink rdfs:Class gnc:ResourceLink)
   (gnc:genemaniaLink rdfs:label "GeneMANIA")
   (gnc:genemaniaLink rdfs:comments "GeneMANIA")
   (gnc:gemmaLink rdfs:Class gnc:ResourceLink)
   (gnc:gemmaLink rdfs:label "Gemma")
   (gnc:gemmaLink rdfs:comments "Meta-analysis of gene expression data")
   (gnc:biogpsLink rdfs:Class gnc:ResourceLink)
   (gnc:biogpsLink rdfs:label "BioGPS")
   (gnc:biogpsLink rdfs:comments "Expression across many tissues and cell types")
   (gnc:abaLink rdfs:Class gnc:ResourceLink)
   (gnc:abaLink rdfs:label "ABA")
   (gnc:abaLink rdfs:comments "Allen Brain Atlas")
   (gnc:ucsRefSeqLink rdfs:Class gnc:ResourceLink)
   (gnc:ucsRefSeqLink rdfs:label "Info from UCSC Genome Browser")
   (gnc:ucsRefSeqLink rdfs:comments "UCSC")
   (gnc:proteinAtlasLink rdfs:Class gnc:ResourceLink)
   (gnc:proteinAtlasLink rdfs:label "Protein Atlas")
   (gnc:proteinAtlasLink rdfs:comments "Human Protein Atlas")
   (gnt:hasChip a owl:ObjectProperty)
   (gnt:hasChip rdfs:domain gnc:Probeset)
   (gnt:hasTargetId a owl:ObjectProperty)
   (gnt:hasTargetId rdfs:domain gnc:Probeset)
   (gnt:symbol rdfs:domain gnc:Probeset)
   (gnt:location rdfs:domain gnc:ProbeSet)
   (gnt:location a owl:ObjectProperty)
   (gnt:strandPosition rdfs:domain gnc:ProbeSet)
   (gnt:strandPosition a owl:ObjectProperty)
   (gnt:targetsRegion a owl:ObjectProperty)
   (gnt:targetsRegion rdfs:domain gnc:Probeset)
   (gnt:chr rdfs:domain gnc:Probeset)
   (gnt:mb rdfs:domain gnc:Probeset)
   (gnt:hasSpecificity a owl:ObjectProperty)
   (gnt:hasSpecificity rdfs:domain gnc:Probeset)
   (gnt:hasBlatScore a owl:ObjectProperty)
   (gnt:hasBlatScore rdfs:domain gnc:Probeset)
   (gnt:hasBlatMbStart a owl:ObjectProperty)
   (gnt:hasBlatMbStart rdfs:domain gnc:Probeset)
   (gnt:hasBlatMbEnd a owl:ObjectProperty)
   (gnt:hasBlatMbEnd rdfs:domain gnc:Probeset)
   (gnt:hasBlatSeq a owl:ObjectProperty)
   (gnt:hasBlatSeq rdfs:domain gnc:Probeset)
   (gnt:hasTargetSeq a owl:ObjectProperty)
   (gnt:hasTargetSeq rdfs:domain gnc:Probeset))
  (triples
      (let ((id (field ("IF(NULLIF(TRIM(ProbeSet.Name), '') IS NULL, '', TRIM(ProbeSet.Name))"
                        ProbeSetIdName)))
            (probeset-id (field ProbeSet Id)))
        (string->identifier
         "probeset"
         (if (string-null? id)
             (number->string probeset-id)
             (regexp-substitute/global
              #f "[^A-Za-z0-9:]"
              id
              'pre "_" 'post))))
    (set rdf:type 'gnc:Probeset)
    (set rdfs:label (field ProbeSet Name))
    (set skos:altLabel
         (replace-substrings
          (field ProbeSet alias)
          '(("\r\n" . "; "))))
    (set gnt:hasChip
         (string->identifier
          "platform"
          (field ("IFNULL(GeneChip.Name, '')" GeneChipName))))
    (set gnt:hasTargetId
         (field ("NULLIF(TRIM(ProbeSet.TargetId), '')"
                 TargetId)))
    (set gnt:symbol (field ProbeSet Symbol))
    (set dct:description (sanitize-rdf-string (field ProbeSet description)))
    (set gnt:targetsRegion
         (sanitize-rdf-string
          (field ("NULLIF(TRIM(ProbeSet.Probe_set_target_region), '')"
                  Probe_set_target_region))))
    (set gnt:chr (field ProbeSet Chr))
    (set gnt:mb (annotate-field (field ("IFNULL(ProbeSet.Mb, '')" Mb)) '^^xsd:double))
    (set gnt:location
         (let* ((mb (field ProbeSet Mb))
                (chr (field ProbeSet Chr))
                (strand-probe (field ProbeSet Strand_Probe))
                (location (list chr mb)))
           (match location
             (("Un" mb)
              (format #f "Not available"))
             ((chr "")
              (if (string-blank? chr)
                  (format #f "Not available")
                  (format #f "Chr ~a @ Unknown position ~a~:[~;~a~]"
                          chr mb
                          (and (string? strand-probe) (or (string=? "+" strand-probe)
                                                          (string=? "-" strand-probe)))
                          (cond ((string=? "+" strand-probe)
                                 " on the plus strand")
                                ((string=? "-" strand-probe)
                                 " on the minus strand")
                                (else "")))))
             (_
              (format #f "Chr ~a @ ~a~:[~;~a~]"
                      chr mb
                      (and (string? strand-probe) (or (string=? "+" strand-probe)
                                                      (string=? "-" strand-probe)))
                      (cond ((string=? "+" strand-probe)
                             " on the plus strand")
                            ((string=? "-" strand-probe)
                             " on the minus strand")
                            (else "")))))))
    ;; NCBI Gene Link
    (set dct:references
         (let ((geneId (field ProbeSet GeneId)))
           (if (not (string-blank? geneId))
               (string->symbol
                (format #f
                        "<~0@*~a~1@*~a> .~%<~0@*~a~1@*~a> ~2@*~a"
                        "http://www.ncbi.nlm.nih.gov/gene?cmd=Retrieve&dopt=Graphics&list_uids="
                        geneId
                        "a gnc:NCBIGeneLink"))
               "")))
    ;; OMIM Link
    (set dct:references
         (let ((omim (field ProbeSet OMIM)))
           (if (not (string-blank? omim))
               (string->symbol
                (format #f
                        "<~0@*~a~1@*~a> .~%<~0@*~a~1@*~a> ~2@*~a"
                        "http://www.ncbi.nlm.nih.gov/omim/"
                        omim
                        "a gnc:omimLink"))
               "")))
    ;; Homologene Link
    (set dct:references
         (let ((homologene (field ProbeSet HomoloGeneID)))
           (if (not (string-blank? homologene))
               (string->symbol
                (format #f
                        "<~0@*~a~1@*~a> .~%<~0@*~a~1@*~a> ~2@*~a"
                        "http://www.ncbi.nlm.nih.gov/homologene/?term="
                        homologene
                        "a gnc:homologeneLink"))
               "")))
    ;; UniProt Link
    (set dct:references
         (let ((uniprot (field ProbeSet UniProtID)))
           (if (not (string-blank? uniprot))
               (string->symbol
                (format #f
                        "<~0@*~a~1@*~a> .~%<~0@*~a~1@*~a> ~2@*~a"
                        "https://www.uniprot.org/uniprot/"
                        uniprot
                        "a gnc:uniprotLink"))
               "")))
    ;; STRING Link
    (set dct:references
         (let ((symbol (field ProbeSet Symbol)))
           (if (not (string-blank? symbol))
               (string->symbol
                (format #f
                        "<~0@*~a~1@*~a> .~%<~0@*~a~1@*~a> ~2@*~a"
                        "http://string-db.org/newstring_cgi/show_network_section.pl?identifier="
                        symbol
                        "a gnc:stringLink"))
               "")))
    ;; GTEX link
    (set dct:references
         (let ((symbol (field ProbeSet Symbol)))
           (if (not (string-blank? symbol))
               (string->symbol
                (format #f
                        "<~0@*~a~1@*~a> .~%<~0@*~a~1@*~a> ~2@*~a"
                        "https://www.gtexportal.org/home/gene/"
                        (string-trim-both symbol)
                        "a gnc:gtexLink"))
               "")))
    ;; EBI GWAS Link
    (set dct:references
         (let ((symbol (field ProbeSet Symbol)))
           (if (not (string-blank? symbol))
               (string->symbol
                (format #f
                        "<~0@*~a~1@*~a> .~%<~0@*~a~1@*~a> ~2@*~a"
                        "https://www.ebi.ac.uk/gwas/search?query="
                        (string-trim-both symbol)
                        "a gnc:ebiGwasLink"))
               "")))
    ;; Protein Atlas Link
    (set dct:references
         (let ((symbol (field ProbeSet Symbol)))
           (if (not (string-blank? symbol))
               (string->symbol
                (format #f
                        "<~0@*~a~1@*~a> .~%<~0@*~a~1@*~a> ~2@*~a"
                        "http://www.proteinatlas.org/search/"
                        (string-trim-both symbol)
                        "a gnc:proteinAtlasLink"))
               "")))
    ;; UCS Link
    (set dct:references
         (let* ((symbol (field ProbeSet Symbol))
                (species (field Species Name))
                (db (if (string=? species "mouse")
                         "mm10" "rn7"))
                (transcriptId (field ProbeSet RefSeq_TranscriptId))
                (kgId (field GeneList_rn33 kgId))
                (transcriptStart
                 (if (string=? species "mouse")
                     (field ("(GeneList.txStart * 1000000)" TranscriptStartMm10))
                     (field ("(GeneList_rn33.txStart * 1000000)" TranscriptStartRn7))))
                (chromosome
                 (if (string=? species "mouse")
                     (field GeneList Chromosome)
                     (field GeneList_rn33 Chromosome)))
                (transcriptEnd
                 (if (string=? species "mouse")
                     (field ("(GeneList.txEnd * 1000000)" TranscriptEndMm10))
                     (field ("(GeneList_rn33.txEnd * 1000000)" TranscriptEndRn7))))
                (url (format
                      #f
                      "http://genome.cse.ucsc.edu/cgi-bin/hgTracks?db=~a&hgg_gene="
                      db)))
           (if (and (not (string-blank? symbol))
                    (not (string-blank? transcriptId))
                    (number? transcriptStart)
                    (number? transcriptEnd)
                    (not (string-blank? chromosome))
                    (or (string=? species "mouse")
                        (string=? species "rat")))
               (string->symbol
                (format #f
                        "<~0@*~a~1@*~a&hgg_chrom=chr~2@*~a&hgg_start=~3@*~a&hgg_end=~4@*~a> .~%<~0@*~a~1@*~a&hgg_chrom=chr~2@*~a&hgg_start=~3@*~a&hgg_end=~4@*~a> ~2@*~a"
                        url
                        transcriptId
                        chromosome
                        transcriptStart
                        transcriptEnd
                        "a gnc:ucsRefSeqLink"
                        ))
               "")))
    ;; PANTHER link
    (set dct:references
         (let ((symbol (field ProbeSet Symbol)))
           (if (not (string-blank? symbol))
               (string->symbol
                (format #f
                        "<~0@*~a~1@*~a> .~%<~0@*~a~1@*~a> ~2@*~a"
                        "http://www.pantherdb.org/genes/geneList.do?searchType=basic&fieldName=all&organism=all&listType=1&fieldValue="
                        (string-trim-both symbol)
                        "a gnc:PantherLink"))
               "")))
    ;; A GeneManiaLink
    (set dct:references
         (let ((symbol (field ProbeSet GeneId))
               (species (lower-case-and-replace-spaces
                         (field Species FullName))))
           (if (and (not (string-blank? symbol))
                    (not (string-blank? species))
                    (or
                     (string=? species "mus-musculus")
                     (string=? species "rattus-norvegicus")
                     (string=? species "homo-sapiens")))
               (string->symbol
                (format #f "<~0@*~a/~1@*~a/~2@*~a> .~%<~0@*~a/~1@*~a/~2@*~a> ~3@*~a"
                        "https://genemania.org/search" species (string-trim-both symbol)
                        "a gnc:genemaniaLink"))
               "")))
    ;; ABA Link
    (set dct:references
         (let ((symbol (field ProbeSet Symbol))
               (geneId (field ProbeSet GeneId))
               (species (field Species name)))
           (if (and (not (string-blank? symbol))
                    (not (string-blank? species))
                    (or (string=? species "human")
                        (string=? species "mouse")))
               (string->symbol
                (format #f "<~0@*~a> .~%<~0@*~a> ~2@*~a"
                        "http://mouse.brain-map.org/search/show?search_type=gene&search_term="
                        "a "
                        (if (string=? species "mouse")
                            (string-trim-both symbol)
                            geneId)))
               "")))
    ;; Gemma Link
    (set dct:references
         (let ((geneId (field ProbeSet GeneId)))
           (if (not (string-blank? geneId))
               (string->symbol
                (format #f
                        "<~0@*~a~1@*~a> .~%<~0@*~a~1@*~a> ~2@*~a"
                        "http://www.chibi.ubc.ca/Gemma/gene/showGene.html?ncbiid="
                        geneId
                        "a gnc:gemmaLink"))
               "")))
    ;; BioGPS Link
    (set dct:references
         (let ((geneId (field ProbeSet GeneId))
               (species (field Species Name)))
           (if (and (not (string-blank? geneId))
                    (not (string-blank? species))
                    (or
                     (string=? species "mouse")
                     (string=? species "rat")
                     (string=? species "humans")))
               (string->symbol
                (format #f
                        "<~0@*~a~1@*~a~2@*~a~3@*~a> .~%<~0@*~a~1@*~a~2@*~a~3@*~a> ~4@*~a ; ~5@*~a"
                        "http://biogps.org/?org="
                        species
                        "#goto=genereport&id="
                        geneId
                        "a gnc:ResourceLink"))
               "")))
    (set gnt:strandProbe
         (field ProbeSet Strand_Probe))
    (set gnt:hasSpecificity
         (field ("IFNULL(ProbeSet.Probe_set_specificity, '')"
                 Probe_set_specificity)))
    (set gnt:hasBlatScore
         (field ("IFNULL(ProbeSet.Probe_set_BLAT_score, '')"
                 Probe_set_BLAT_score)))
    (set gnt:hasBlatMbStart
         (annotate-field (field ("IFNULL(ProbeSet.Probe_set_Blat_Mb_start, '')"
                                 Probe_set_Blat_Mb_start))
                         '^^xsd:double))
    (set gnt:hasBlatMbEnd
         (annotate-field (field ("IFNULL(ProbeSet.Probe_set_Blat_Mb_end, '')"
                                 Probe_set_Blat_Mb_end))
                         '^^xsd:double))
    (set gnt:hasBlatSeq (sanitize-rdf-string (field ProbeSet BlatSeq)))
    (set gnt:hasTargetSeq (sanitize-rdf-string (field ProbeSet TargetSeq)))))




(let* ((option-spec
        '((settings (single-char #\s) (value #t))
          (output (single-char #\o) (value #t))
          (documentation (single-char #\d) (value #t))))
       (options (getopt-long (command-line) option-spec))
       (settings (option-ref options 'settings #f))
       (output (option-ref options 'output #f))
       (documentation (option-ref options 'documentation #f))
       (%connection-settings
        (call-with-input-file settings
          read)))
  (with-documentation
   (name "ProbeSet Metadata")
   (connection %connection-settings)
   (table-metadata? #f)
   (prefixes
    '(("gn:" "<http://genenetwork.org/id/>")
      ("probeset:" "<http://genenetwork.org/probeset/>")
      ("gnc:" "<http://genenetwork.org/category/>")
      ("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#>")
      ("dct:" "<http://purl.org/dc/terms/>")
      ("owl:" "<http://www.w3.org/2002/07/owl#>")
      ("xsd:" "<http://www.w3.org/2001/XMLSchema#>")
      ("qb:" "<http://purl.org/linked-data/cube#>")
      ("sdmx-measure:" "<http://purl.org/linked-data/sdmx/2009/measure#>")
      ("skos:" "<http://www.w3.org/2004/02/skos/core#>")))
   (inputs
    (list probeset))
   (outputs
    `(#:documentation ,documentation
      #:rdf ,output))))