From 02c0b2604b3161db5152faafe5bb140fbd61de2d Mon Sep 17 00:00:00 2001
From: Munyoki Kilyungi
Date: Fri, 30 Jan 2026 00:54:27 +0300
Subject: Add phenotype dataset metadata.

Signed-off-by: Munyoki Kilyungi <me@bonfacemunyoki.com>
---
 examples/phenotype-datasets.scm | 74 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 74 insertions(+)
 create mode 100755 examples/phenotype-datasets.scm

diff --git a/examples/phenotype-datasets.scm b/examples/phenotype-datasets.scm
new file mode 100755
index 0000000..9354926
--- /dev/null
+++ b/examples/phenotype-datasets.scm
@@ -0,0 +1,74 @@
+#! /usr/bin/env guile
+!#
+
+(use-modules (rnrs programs)
+             (rnrs io ports)
+             (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 gn:set->gn:dataset
+  (tables (Datasets
+           (inner-join InfoFiles "ON InfoFiles.DatasetId = Datasets.DatasetId")
+           (inner-join InbredSet "ON InbredSet.Id = InfoFiles.InbredSetId")
+           (inner-join PublishFreeze "ON PublishFreeze.InbredSetId = InbredSet.Id"))
+          "WHERE PublishFreeze.public > 0 GROUP BY Datasets.DatasetId")
+  (schema-triples
+   (gnt:has_phenotype_data rdf:type owl:ObjectProperty)
+   (gnt:has_phenotype_data rdfs:label "this resources has this phenotype data.")
+   (gnt:has_phenotype_data rdfs:comment "Associates a resource with its phenotype data.")
+   (gnt:has_phenotype_data rdfs:domain gnc:set)
+   (gnt:has_phenotype_data rdfs:range dcat:Dataset)
+   (gnt:has_phenotype_data rdfs:subPropertyOf dct:relation))
+  (triples (string->identifier "set" (field InbredSet Name InbredSetName) #:separator "_")
+    (set gnt:has_phenotype_data (string->identifier "dataset" (field PublishFreeze Name) #:separator "_"))))
+
+(define-transformer gn:dataset->gn:set
+  (tables (Datasets
+           (inner-join InfoFiles "ON InfoFiles.DatasetId = Datasets.DatasetId")
+           (inner-join InbredSet "ON InbredSet.Id = InfoFiles.InbredSetId")
+           (inner-join PublishFreeze "ON PublishFreeze.InbredSetId = InbredSet.Id"))
+          "WHERE PublishFreeze.public > 0 GROUP BY Datasets.DatasetId")
+  (triples (string->identifier "dataset" (field PublishFreeze Name) #:separator "_")
+    (set gnt:has_strain (string->identifier "set" (field InbredSet Name InbredSetName) #:separator "_"))))
+
+
+(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 "Genotype Datasets")
+   (connection %connection-settings)
+   (table-metadata? #f)
+   (prefixes
+    '(("dcat:" "<http://www.w3.org/ns/dcat#>")
+      ("dct:" "<http://purl.org/dc/terms/>")
+      ("gn:" "<http://rdf.genenetwork.org/v1/id/>")
+      ("gnc:" "<http://rdf.genenetwork.org/v1/category/>")
+      ("gnt:" "<http://rdf.genenetwork.org/v1/term/>")
+      ("rdf:" "<http://www.w3.org/1999/02/22-rdf-syntax-ns#>")
+      ("rdfs:" "<http://www.w3.org/2000/01/rdf-schema#>")
+      ("owl:" "<http://www.w3.org/2002/07/owl#>")
+      ("skos:" "<http://www.w3.org/2004/02/skos/core#>")
+      ("xkos:" "<http://rdf-vocabulary.ddialliance.org/xkos#>")
+      ("xsd:" "<http://www.w3.org/2001/XMLSchema#>")))
+   (inputs
+    (list gn:set->gn:dataset gn:dataset->gn:set))
+   (outputs
+    `(#:documentation ,documentation
+      #:rdf ,output))))
-- 
cgit 1.4.1