fix: adapter review findings (BrAPI/ISA-Tab/PRIDE/ENA correctness)

src/metaseed/brapi/export.py

@@ -73,22 +73,34 @@ def _observation_unit(unit: dict[str, Any]) -> dict[str, Any]:
             "levelCode": unit.get("observation_level_code"),
         }
     )
+    # BrAPI v2: block/replicate are expressed via observationLevelRelationships.
+    relationships = []
+    if unit.get("observation_unit_block"):
+        relationships.append(
+            {"levelName": "block", "levelCode": str(unit["observation_unit_block"])}
+        )
+    if unit.get("observation_unit_replicate"):
+        relationships.append(
+            {"levelName": "rep", "levelCode": str(unit["observation_unit_replicate"])}
+        )
     position = _clean(
         {
             "positionCoordinateX": unit.get("observation_unit_x_ref"),
             "positionCoordinateXType": unit.get("spatial_distribution_type"),
             "positionCoordinateY": unit.get("observation_unit_y_ref"),
-            "blockNumber": unit.get("observation_unit_block"),
-            "replicate": unit.get("observation_unit_replicate"),
             "entryType": unit.get("entry_type"),
         }
     )
+    # BrAPI v2 nests the level inside the position object.
+    if level:
+        position["observationLevel"] = level
+    if relationships:
+        position["observationLevelRelationships"] = relationships
     return _clean(
         {
             "observationUnitDbId": unit.get("unique_id"),
             "studyDbId": unit.get("study_id"),
             "germplasmDbId": unit.get("biological_material_id"),
-            "observationLevel": level,
             "observationUnitPosition": position,
         }
     )

src/metaseed/brapi/mapper.py

@@ -156,8 +156,10 @@ def _add_observation_units(
         unit_id = unit.get("observationUnitDbId")
         if not unit_id:
             continue
-        level = unit.get("observationLevel") or {}
         position = unit.get("observationUnitPosition") or {}
+        # In BrAPI v2 the observationLevel object is nested inside the position,
+        # and block/replicate are expressed via observationLevelRelationships.
+        level = position.get("observationLevel") or {}
         client.create_entity(
             "ObservationUnit",
             _clean(
@@ -173,15 +175,26 @@ def _add_observation_units(
                     ),
                     "observation_unit_x_ref": position.get("positionCoordinateX"),
                     "observation_unit_y_ref": position.get("positionCoordinateY"),
-                    "observation_unit_block": position.get("blockNumber"),
-                    "observation_unit_replicate": position.get("replicate"),
+                    "observation_unit_block": _level_code(position, ("block",)),
+                    "observation_unit_replicate": _level_code(
+                        position, ("rep", "replicate")
+                    ),
                     "entry_type": position.get("entryType"),
                 }
             ),
             skip_validation=True,
         )
 
 
+def _level_code(position: dict[str, Any], level_names: tuple[str, ...]) -> str | None:
+    """Return the levelCode for a named level from observationLevelRelationships."""
+    for rel in position.get("observationLevelRelationships") or []:
+        if str(rel.get("levelName", "")).lower() in level_names:
+            code = rel.get("levelCode")
+            return str(code) if code is not None else None
+    return None
+
+
 def _add_observed_variables(
     client: MetaseedClient,
     observations: list[dict[str, Any]],

src/metaseed/ena/mapper.py

@@ -14,6 +14,7 @@
 
 from __future__ import annotations
 
+from itertools import zip_longest
 from typing import TYPE_CHECKING, Any
 
 if TYPE_CHECKING:
@@ -142,9 +143,14 @@ def build_dataset(
             skip_validation=True,
         )
 
-        urls = [u for u in (row.get("fastq_ftp") or "").split(";") if u]
+        # fastq_ftp and fastq_md5 are parallel ;-separated lists; pair them
+        # positionally (zip_longest) so an empty URL segment cannot shift the
+        # checksums out of alignment.
+        urls = (row.get("fastq_ftp") or "").split(";")
         md5s = (row.get("fastq_md5") or "").split(";")
-        for i, url in enumerate(urls):
+        for url, md5 in zip_longest(urls, md5s, fillvalue=""):
+            if not url:
+                continue
             client.create_entity(
                 "File",
                 _clean(
@@ -153,7 +159,7 @@ def build_dataset(
                         "filename": url.rsplit("/", 1)[-1],
                         "filetype": "fastq",
                         "checksum_method": "MD5",
-                        "checksum": md5s[i] if i < len(md5s) else None,
+                        "checksum": md5 or None,
                     }
                 ),
                 skip_validation=True,

src/metaseed/isatab/__init__.py

@@ -7,9 +7,10 @@
 
 This writer is pure and dependency-free (stdlib tab-delimited text). It emits
 the metadata structure and *references* raw data files by name; it never reads
-or writes spectra. It works for any ISA-shaped profile (e.g. ``isa`` or
-``metabolights``) whose serialized entities use the ISA entity types
-Investigation, Study, Person, Publication, Factor, Protocol, and Assay.
+or writes spectra. It reads the ISA entity types Investigation, Study, Person,
+Publication, Factor, Protocol, and Assay using the field codenames of the
+``metabolights`` profile; other ISA-shaped profiles (e.g. ``isa``) work only
+where their field codenames match.
 """
 
 from __future__ import annotations
@@ -221,20 +222,61 @@ def to_isatab(client: MetaseedClient) -> dict[str, str]:
     assays = by_type.get("Assay", [])
     samples = by_type.get("Sample", [])
 
+    # Each entity's owning study (None = directly under the Investigation), so a
+    # multi-study investigation routes factors/protocols/assays/samples and
+    # study- vs investigation-level people/publications to the right place
+    # instead of duplicating every child into every study.
+    owner = _parent_study_map(client)
+
+    def of(items: list[dict[str, Any]], study_id: str | None) -> list[dict[str, Any]]:
+        return [e for e in items if owner.get(str(e.get("_node_id"))) == study_id]
+
     lines = _ontology_source_section()
     investigation = investigations[0] if investigations else {}
     lines += _investigation_section(investigation)
-    lines += _publication_section("INVESTIGATION PUBLICATIONS", "Investigation", [])
-    lines += _contacts_section("INVESTIGATION CONTACTS", "Investigation", [])
+    lines += _publication_section(
+        "INVESTIGATION PUBLICATIONS", "Investigation", of(publications, None)
+    )
+    lines += _contacts_section(
+        "INVESTIGATION CONTACTS", "Investigation", of(people, None)
+    )
 
     for study in studies:
+        sid = study.get("_node_id")
         lines += _study_sections(
-            study, factors, assays, protocols, publications, people
+            study,
+            of(factors, sid),
+            of(assays, sid),
+            of(protocols, sid),
+            of(publications, sid),
+            of(people, sid),
         )
 
     documents: dict[str, str] = {"i_Investigation.txt": "\n".join(lines) + "\n"}
     for study in studies:
-        documents[study_filename(study)] = _study_file(samples)
+        documents[study_filename(study)] = _study_file(
+            of(samples, study.get("_node_id"))
+        )
     for assay in assays:
         documents[assay_filename(assay)] = _assay_file(assay)
     return documents
+
+
+def _parent_study_map(client: MetaseedClient) -> dict[str, str | None]:
+    """Map each entity's node id to the node id of the Study that owns it.
+
+    Entities directly under the Investigation (e.g. investigation-level people
+    or publications) map to ``None``. Built from the dataset tree, since the flat
+    serialization does not carry parent links.
+    """
+    owner: dict[str, str | None] = {}
+
+    def descend(node: Any, study_id: str | None) -> None:
+        for child in node.children:
+            owner[child.id] = study_id
+            next_study = child.id if child.entity_type == "Study" else study_id
+            descend(child, next_study)
+
+    for root in client.get_tree():
+        descend(root, None)
+    return owner

src/metaseed/metabolights/export.py

@@ -1,7 +1,7 @@
 """Export a ``metabolights`` dataset to the MetaboLights submission format.
 
 MetaboLights archives are ISA-Tab plus one MetaboLights-specific file per
-assay: the Metabolite Assignment File (MAF, ``m_*.txt``). This module reuses the
+assay: the Metabolite Assignment File (MAF, ``m_*.tsv``). This module reuses the
 shared :func:`metaseed.isatab.to_isatab` writer for the ISA-Tab documents and
 adds a MAF skeleton (the standard column header) for each assay. Pure and
 dependency-free; it references files, never reads or writes spectra.
@@ -57,7 +57,7 @@ def to_metabolights(client: MetaseedClient) -> dict[str, str]:
 
     Returns:
         Mapping of document name to text — the ISA-Tab files from
-        :func:`metaseed.isatab.to_isatab` plus one ``m_*.txt`` MAF (column
+        :func:`metaseed.isatab.to_isatab` plus one ``m_*.tsv`` MAF (column
         header only) per Assay.
     """
     documents = dict(to_isatab(client))

src/metaseed/pride/export.py

@@ -52,6 +52,8 @@ def _metadata_lines(dataset: dict[str, Any]) -> list[str]:
         _mtd("submitter_email", submitter.get("email")),
         _mtd("submitter_affiliation", submitter.get("affiliation")),
         _mtd("lab_head_name", lab_head.get("name")),
+        _mtd("lab_head_email", lab_head.get("email")),
+        _mtd("lab_head_affiliation", lab_head.get("affiliation")),
         _mtd("project_title", dataset.get("title")),
         _mtd("project_description", dataset.get("description")),
         _mtd("keywords", ", ".join(str(k) for k in keywords)),
@@ -68,18 +70,19 @@ def _metadata_lines(dataset: dict[str, Any]) -> list[str]:
 
 
 def _file_lines(dataset: dict[str, Any]) -> list[str]:
-    """Render one ``FME`` file-mapping entry per referenced data file."""
-    lines = []
-    for index, file in enumerate(dataset.get("files") or []):
-        filename = file.get("filename")
-        if not filename:
-            continue
-        file_type = file.get("file_type") or ""
-        fields = [str(index), str(filename), str(file_type)]
-        file_format = file.get("file_format")
-        if file_format:
-            fields.append(str(file_format))
-        lines.append("FME\t" + "\t".join(fields))
+    """Render the ``FMH`` header and one ``FME`` entry per referenced data file.
+
+    The PRIDE px file-mapping section is an ``FMH`` header row declaring the
+    columns followed by ``FME`` rows ordered ``file_id, file_type, file_path``.
+    """
+    files = [f for f in (dataset.get("files") or []) if f.get("filename")]
+    if not files:
+        return []
+    lines = ["FMH\tfile_id\tfile_type\tfile_path"]
+    for file_id, file in enumerate(files):
+        file_type = str(file.get("file_type") or "")
+        path = str(file.get("filename"))
+        lines.append(f"FME\t{file_id}\t{file_type}\t{path}")
     return lines
 
 

tests/test_brapi/fixtures/brapi.json

@@ -1,6 +1,12 @@
 {
   "studies": {
-    "metadata": {"pagination": {"currentPage": 0, "totalCount": 2, "totalPages": 1}},
+    "metadata": {
+      "pagination": {
+        "currentPage": 0,
+        "totalCount": 2,
+        "totalPages": 1
+      }
+    },
     "result": {
       "data": [
         {
@@ -12,8 +18,13 @@
           "startDate": "2024-04-01",
           "endDate": "2024-09-30",
           "locationName": "Wageningen field A",
-          "experimentalDesign": {"PUI": "CO_715:0000145", "description": "RCBD"},
-          "growthFacility": {"description": "open field"},
+          "experimentalDesign": {
+            "PUI": "CO_715:0000145",
+            "description": "RCBD"
+          },
+          "growthFacility": {
+            "description": "open field"
+          },
           "documentationURL": "https://example.org/studies/1001",
           "dataLinks": [
             {
@@ -38,7 +49,13 @@
     }
   },
   "observationunits": {
-    "metadata": {"pagination": {"currentPage": 0, "totalCount": 2, "totalPages": 1}},
+    "metadata": {
+      "pagination": {
+        "currentPage": 0,
+        "totalCount": 2,
+        "totalPages": 1
+      }
+    },
     "result": {
       "data": [
         {
@@ -47,14 +64,25 @@
           "studyDbId": "1001",
           "germplasmDbId": "G1",
           "germplasmName": "B73",
-          "observationLevel": {"levelName": "plot", "levelCode": "1"},
           "observationUnitPosition": {
             "positionCoordinateX": "1",
             "positionCoordinateXType": "GRID_COL",
             "positionCoordinateY": "1",
-            "blockNumber": "1",
-            "replicate": "1",
-            "entryType": "TEST"
+            "entryType": "TEST",
+            "observationLevel": {
+              "levelName": "plot",
+              "levelCode": "1"
+            },
+            "observationLevelRelationships": [
+              {
+                "levelName": "block",
+                "levelCode": "1"
+              },
+              {
+                "levelName": "rep",
+                "levelCode": "1"
+              }
+            ]
           }
         },
         {
@@ -63,21 +91,38 @@
           "studyDbId": "1001",
           "germplasmDbId": "G2",
           "germplasmName": "Mo17",
-          "observationLevel": {"levelName": "plot", "levelCode": "2"},
           "observationUnitPosition": {
             "positionCoordinateX": "2",
             "positionCoordinateXType": "GRID_COL",
             "positionCoordinateY": "1",
-            "blockNumber": "1",
-            "replicate": "2",
-            "entryType": "TEST"
+            "entryType": "TEST",
+            "observationLevel": {
+              "levelName": "plot",
+              "levelCode": "2"
+            },
+            "observationLevelRelationships": [
+              {
+                "levelName": "block",
+                "levelCode": "1"
+              },
+              {
+                "levelName": "rep",
+                "levelCode": "2"
+              }
+            ]
           }
         }
       ]
     }
   },
   "observations": {
-    "metadata": {"pagination": {"currentPage": 0, "totalCount": 3, "totalPages": 1}},
+    "metadata": {
+      "pagination": {
+        "currentPage": 0,
+        "totalCount": 3,
+        "totalPages": 1
+      }
+    },
     "result": {
       "data": [
         {
@@ -108,7 +153,13 @@
     }
   },
   "germplasm": {
-    "metadata": {"pagination": {"currentPage": 0, "totalCount": 2, "totalPages": 1}},
+    "metadata": {
+      "pagination": {
+        "currentPage": 0,
+        "totalCount": 2,
+        "totalPages": 1
+      }
+    },
     "result": {
       "data": [
         {
@@ -120,7 +171,9 @@
           "subtaxa": "subsp. mays",
           "instituteCode": "USDA",
           "instituteName": "USDA-ARS",
-          "studyDbIds": ["1001"]
+          "studyDbIds": [
+            "1001"
+          ]
         },
         {
           "germplasmDbId": "G2",