cleaning up biosample and participant

dandi/metadata.py

@@ -228,6 +228,8 @@ def extract_anatomy(metadata):
 def extract_model(modelcls, metadata, **kwargs):
     m = modelcls.unvalidated()
     for field in m.__fields__.keys():
+        if modelcls == models.BioSample and field == "wasDerivedFrom":
+            continue
         value = kwargs.get(field, extract_field(field, metadata))
         if value is not Ellipsis:
             setattr(m, field, value)
@@ -237,7 +239,17 @@ def extract_model(modelcls, metadata, **kwargs):
 
 def extract_wasDerivedFrom(metadata):
     return [
-        extract_model(models.BioSample, metadata, identifier=metadata.get("subject_id"))
+        extract_model(
+            models.BioSample, metadata, identifier=metadata.get("tissue_sample_id")
+        )
+    ]
+
+
+def extract_wasAttributedTo(metadata):
+    return [
+        extract_model(
+            models.Participant, metadata, identifier=metadata.get("subject_id")
+        )
     ]
 
 
@@ -253,6 +265,7 @@ def extract_digest(metadata):
 
 FIELD_EXTRACTORS = {
     "wasDerivedFrom": extract_wasDerivedFrom,
+    "wasAttributedTo": extract_wasAttributedTo,
     "age": extract_age,
     "sex": extract_sex,
     "assayType": extract_assay_type,
@@ -348,7 +361,7 @@ def toContributor(value):
                     roles.append("".join([val.capitalize() for val in tmp]))
                 else:
                     roles.append(tmp.pop())
-            contrib["roleName"] = roles
+            contrib["roleName"] = [getattr(models.RoleType, role) for role in roles]
             del item["roles"]
         if "awardNumber" in item:
             contrib["awardNumber"] = item["awardNumber"]
@@ -362,9 +375,17 @@ def toContributor(value):
                 contrib["identifier"] = models.PropertyValue()
             del item["orcid"]
         if "affiliations" in item:
-            item["affiliation"] = item["affiliations"]
+            item["affiliation"] = [
+                models.Organization.unvalidated(**{"name": affiliate})
+                for affiliate in item["affiliations"]
+            ]
+
             del item["affiliations"]
         contrib.update(**{f"{k}": v for k, v in item.items()})
+        if "awardNumber" in contrib:
+            contrib = models.Organization.unvalidated(**contrib)
+        else:
+            contrib = models.Person.unvalidated(**contrib)
         out.append(contrib)
     return out
 
@@ -404,14 +425,18 @@ def convertv1(data):
                 out = []
                 for item in value:
                     if isinstance(item, dict):
-                        out.append({k: v for k, v in item.items()})
+                        out.append(
+                            models.Resource.unvalidated(
+                                **{k: v for k, v in item.items()}
+                            )
+                        )
                     else:
                         present = False
                         for val in out:
                             if item in val.values():
                                 present = True
                         if not present:
-                            out.append({"url": item})
+                            out.append(models.Resource.unvalidated(**{"url": item}))
                 value = out
             if oldkey in [
                 "number_of_subjects",
@@ -423,7 +448,14 @@ def convertv1(data):
             if isinstance(value, list):
                 for val in value:
                     if extrakey:
-                        val[extrakey] = extra
+                        if extrakey == "relation":
+                            val.relation = getattr(models.RelationType, extra)
+                        elif extrakey == "roleName":
+                            val.roleName = [
+                                getattr(models.RoleType, role) for role in extra
+                            ]
+                        else:
+                            val[extrakey] = extra
             if isinstance(value, dict):
                 if extrakey:
                     value[extrakey] = extra

dandi/models.py

@@ -226,17 +226,17 @@ class ContactPoint(DandiBaseModel):
 
 
 class Contributor(DandiBaseModel):
-    identifier: Identifier = Field(
+    identifier: Optional[Identifier] = Field(
         None,
         title="A Common Identifier",
         description="Use a common identifier such as ORCID for people or ROR for institutions",
         nskey="schema",
     )
-    name: str = Field(None, nskey="schema")
-    email: EmailStr = Field(None, nskey="schema")
-    url: AnyUrl = Field(None, nskey="schema")
-    roleName: List[RoleType] = Field(
-        title="Role", description="Role of the contributor", nskey="schema"
+    name: Optional[str] = Field(None, nskey="schema")
+    email: Optional[EmailStr] = Field(None, nskey="schema")
+    url: Optional[AnyUrl] = Field(None, nskey="schema")
+    roleName: Optional[List[RoleType]] = Field(
+        None, title="Role", description="Role of the contributor", nskey="schema"
     )
     includeInCitation: bool = Field(
         True,
@@ -245,7 +245,7 @@ class Contributor(DandiBaseModel):
         "when generating a citation for the item",
         nskey="dandi",
     )
-    awardNumber: Identifier = Field(
+    awardNumber: Optional[Identifier] = Field(
         None,
         title="Identifier for an award",
         description="Identifier associated with a sponsored or gidt award",
@@ -254,8 +254,15 @@ class Contributor(DandiBaseModel):
 
 
 class Organization(Contributor):
-    contactPoint: List[ContactPoint] = Field(
-        description="Contact for the organization", nskey="schema"
+    includeInCitation: bool = Field(
+        False,
+        title="Include Contributor in Citation",
+        description="A flag to indicate whether a contributor should be included "
+        "when generating a citation for the item",
+        nskey="dandi",
+    )
+    contactPoint: Optional[List[ContactPoint]] = Field(
+        None, description="Contact for the organization", nskey="schema"
     )
     _ldmeta = {
         "rdfs:subClassOf": ["schema:Organization", "prov:Organization"],
@@ -396,9 +403,29 @@ class BioSample(DandiBaseModel):
     )
     anatomy: Optional[List[Anatomy]] = Field(
         None,
-        description="UBERON based identifier for the location of the sample",
+        description="UBERON based identifier for what organ the sample belongs "
+        "to. Use the most specific descriptor.",
         nskey="dandi",
     )
+
+    wasDerivedFrom: Optional[List["BioSample"]] = Field(None, nskey="prov")
+
+    _ldmeta = {
+        "rdfs:subClassOf": ["schema:Thing", "prov:Entity"],
+        "rdfs:label": "Information about the biosample.",
+        "nskey": "dandi",
+    }
+
+
+BioSample.update_forward_refs()
+
+
+class Participant(DandiBaseModel):
+    """Description about the sample that was studied"""
+
+    identifier: Identifier = Field(nskey="schema")
+    source_id: Optional[Identifier] = Field(None, nskey="dandi")
+
     strain: Optional[StrainType] = Field(
         None, description="Identifier for the strain of the sample", nskey="dandi"
     )
@@ -435,7 +462,7 @@ class BioSample(DandiBaseModel):
 
     _ldmeta = {
         "rdfs:subClassOf": ["schema:Thing", "prov:Entity"],
-        "rdfs:label": "Information about the biosample.",
+        "rdfs:label": "Information about the participant.",
         "nskey": "dandi",
     }
 
@@ -632,6 +659,9 @@ class AssetMeta(CommonModel):
     )
 
     wasDerivedFrom: Optional[List[BioSample]] = Field(None, nskey="prov")
+    wasAttributedTo: List[Participant] = Field(
+        None, description="Participant(s) to which this file belongs to", nskey="prov"
+    )
 
     # on publish or set by server
     contentUrl: Optional[List[AnyUrl]] = Field(None, readOnly=True, nskey="schema")