Merge pull request #2107 from effigies/schema/phenotype-specificity

feat(schema): Rewrite PHENOTYPE_SUBJECTS_MISSING to run on each phenotype file

Author: effigies

src/schema/meta/context.yaml

@@ -80,11 +80,6 @@ properties:
             type: array
             items:
               type: string
-          phenotype:
-            description: 'The union of participant_id columns in phenotype files'
-            type: array
-            items:
-              type: string
   subject:
     description: 'Properties and contents of the current subject'
     type: object
@@ -109,11 +104,6 @@ properties:
             type: array
             items:
               type: string
-          phenotype:
-            description: 'The union of session_id columns in phenotype files'
-            type: array
-            items:
-              type: string
 
   # Properties of the current file
   path:

src/schema/rules/checks/dataset.yaml

@@ -30,24 +30,6 @@ ParticipantIDMismatch:
         sorted(dataset.subjects.sub_dirs)
       )
 
-# 51
-PhenotypeSubjectsMissing:
-  issue:
-    code: PHENOTYPE_SUBJECTS_MISSING
-    message: |
-      A phenotype/ .tsv file lists subjects that were not found in
-      the participant_id column found in the participants.tsv file.
-    level: error
-  selectors:
-    - path == '/participants.tsv'
-    - type(dataset.subjects.phenotype) != 'null'
-  checks:
-    - |
-      allequal(
-        sorted(intersects(columns.participant_id, dataset.subjects.phenotype)),
-        sorted(dataset.subjects.phenotype)
-      )
-
 # 214
 SamplesTSVMissing:
   issue:

src/schema/rules/checks/phenotype.yaml

@@ -0,0 +1,21 @@
+# Rules for phenotype files
+---
+
+# 51
+PhenotypeSubjectsMissing:
+  issue:
+    code: PHENOTYPE_SUBJECTS_MISSING
+    message: |
+      A phenotype/ .tsv file lists subjects that were not found in
+      the participant_id column found in the participants.tsv file.
+    level: error
+  selectors:
+    - datatype == 'phenotype'
+    - suffix == '.tsv'
+    - type(dataset.subjects.participant_id) != 'null'
+  checks:
+    - |
+      allequal(
+        sorted(intersects(columns.participant_id, dataset.subjects.participant_id)),
+        sorted(dataset.subjects.participant_id)
+      )