nmdc-runtime 2.2.1__py3-none-any.whl → 2.4.0__py3-none-any.whl

nmdc_runtime/site/repair/database_updater.py

@@ -0,0 +1,230 @@
+from functools import lru_cache
+from typing import Any, Dict, List
+import pandas as pd
+from nmdc_runtime.site.resources import (
+    RuntimeApiUserClient,
+    RuntimeApiSiteClient,
+    GoldApiClient,
+)
+from nmdc_runtime.site.translation.gold_translator import GoldStudyTranslator
+from nmdc_schema import nmdc
+
+
+class DatabaseUpdater:
+    def __init__(
+        self,
+        runtime_api_user_client: RuntimeApiUserClient,
+        runtime_api_site_client: RuntimeApiSiteClient,
+        gold_api_client: GoldApiClient,
+        study_id: str,
+        gold_nmdc_instrument_map_df: pd.DataFrame = pd.DataFrame(),
+    ):
+        """This class serves as an API for repairing connections in the database by
+        adding records that are essentially missing "links"/"connections". As we identify
+        common use cases for adding missing records to the database, we can
+        add helper methods to this class.
+
+        :param runtime_api_user_client: An object of RuntimeApiUserClient which can be
+        used to retrieve instance records from the NMDC database.
+        :param runtime_api_site_client: An object of RuntimeApiSiteClient which can be
+        used to mint new IDs for the repaired records that need to be added into the NMDC database.
+        :param gold_api_client: An object of GoldApiClient which can be used to retrieve
+        records from GOLD via the GOLD API.
+        :param study_id: NMDC study ID for which the missing records need to be added.
+        :param gold_nmdc_instrument_map_df: A dataframe originally stored as a TSV mapping file in the
+        NMDC schema repo, which maps GOLD instrument IDs to IDs of NMDC instrument_set records.
+        """
+        self.runtime_api_user_client = runtime_api_user_client
+        self.runtime_api_site_client = runtime_api_site_client
+        self.gold_api_client = gold_api_client
+        self.study_id = study_id
+        self.gold_nmdc_instrument_map_df = gold_nmdc_instrument_map_df
+
+    @lru_cache
+    def _fetch_gold_biosample(self, gold_biosample_id: str) -> List[Dict[str, Any]]:
+        """Fetch response from GOLD /biosamples API for a given biosample id.
+
+        :param gold_biosample_id: GOLD biosample ID.
+        :return: Dictionary containing the response from the GOLD /biosamples API.
+        """
+        return self.gold_api_client.fetch_biosample_by_biosample_id(gold_biosample_id)
+
+    @lru_cache
+    def _fetch_gold_projects(self, gold_biosample_id: str):
+        """Fetch response from GOLD /projects API for a given biosample id.
+
+        :param gold_biosample_id: GOLD biosample ID
+        :return: Dictionary containing the response from the GOLD /projects API.
+        """
+        return self.gold_api_client.fetch_projects_by_biosample(gold_biosample_id)
+
+    def generate_data_generation_set_records_from_gold_api_for_study(
+        self,
+    ) -> nmdc.Database:
+        """This method creates missing data generation records for a given study in the NMDC database using
+        metadata from GOLD. The way the logic works is, it first fetches all the biosamples associated
+        with the study from the NMDC database. Then, it fetches all the biosample and project data data
+        associated with the individual biosamples from the GOLD API using the NMDC-GOLD biosample id
+        mappings on the "gold_biosample_identifiers" key/slot. We use the GoldStudyTranslator class
+        to mint the required number of `nmdc:DataGeneration` (`nmdc:NucleotideSequencing`) records based
+        on the number of GOLD sequencing projects, and then reimplement only the part of logic from that
+        class which is responsible for making data_generation_set records.
+
+        :return: An instance of `nmdc:Database` object which is JSON-ified and rendered on the frontend.
+        """
+        database = nmdc.Database()
+
+        biosample_set = self.runtime_api_user_client.get_biosamples_for_study(
+            self.study_id
+        )
+
+        all_gold_biosamples = []
+        all_gold_projects = []
+        for biosample in biosample_set:
+            gold_biosample_identifiers = biosample.get("gold_biosample_identifiers")
+            if gold_biosample_identifiers:
+                for gold_biosample_id in gold_biosample_identifiers:
+                    gold_biosample = self._fetch_gold_biosample(gold_biosample_id)[0]
+                    gold_projects = self._fetch_gold_projects(gold_biosample_id)
+                    gold_biosample["projects"] = gold_projects
+
+                    all_gold_biosamples.append(gold_biosample)
+                    all_gold_projects.extend(gold_projects)
+
+        gold_study_translator = GoldStudyTranslator(
+            biosamples=all_gold_biosamples,
+            projects=all_gold_projects,
+            gold_nmdc_instrument_map_df=self.gold_nmdc_instrument_map_df,
+        )
+
+        # The GoldStudyTranslator class has some pre-processing logic which filters out
+        # invalid biosamples and projects (based on `sequencingStrategy`, `projectStatus`, etc.)
+        filtered_biosamples = gold_study_translator.biosamples
+        filtered_projects = gold_study_translator.projects
+
+        gold_project_ids = [project["projectGoldId"] for project in filtered_projects]
+        nmdc_nucleotide_sequencing_ids = self.runtime_api_site_client.mint_id(
+            "nmdc:NucleotideSequencing", len(gold_project_ids)
+        ).json()
+        gold_project_to_nmdc_nucleotide_sequencing_ids = dict(
+            zip(gold_project_ids, nmdc_nucleotide_sequencing_ids)
+        )
+
+        gold_to_nmdc_biosample_ids = {}
+
+        for biosample in biosample_set:
+            gold_ids = biosample.get("gold_biosample_identifiers", [])
+            for gold_id in gold_ids:
+                gold_id_stripped = gold_id.replace("gold:", "")
+                gold_to_nmdc_biosample_ids[gold_id_stripped] = biosample["id"]
+
+        database.data_generation_set = []
+        # Similar to the logic in GoldStudyTranslator, the number of nmdc:NucleotideSequencing records
+        # created is based on the number of GOLD sequencing projects
+        for project in filtered_projects:
+            # map the projectGoldId to the NMDC biosample ID
+            biosample_gold_id = next(
+                (
+                    biosample["biosampleGoldId"]
+                    for biosample in filtered_biosamples
+                    if any(
+                        p["projectGoldId"] == project["projectGoldId"]
+                        for p in biosample.get("projects", [])
+                    )
+                ),
+                None,
+            )
+
+            if biosample_gold_id:
+                nmdc_biosample_id = gold_to_nmdc_biosample_ids.get(biosample_gold_id)
+                if nmdc_biosample_id:
+                    database.data_generation_set.append(
+                        gold_study_translator._translate_nucleotide_sequencing(
+                            project,
+                            nmdc_nucleotide_sequencing_id=gold_project_to_nmdc_nucleotide_sequencing_ids[
+                                project["projectGoldId"]
+                            ],
+                            nmdc_biosample_id=nmdc_biosample_id,
+                            nmdc_study_id=self.study_id,
+                        )
+                    )
+
+        return database
+
+    def generate_biosample_set_from_gold_api_for_study(self) -> nmdc.Database:
+        """This method creates biosample_set records for a given study in the NMDC database using
+        metadata from GOLD. The logic works by first fetching the biosampleGoldId values of all
+        biosamples associated with the study. Then, it fetches the list of all biosamples associated
+        with the GOLD study using the GOLD API. There's pre-processing logic in the GoldStudyTranslator
+        to filter out biosamples based on `sequencingStrategy` and `projectStatus`. On this list of
+        filtered biosamples, we compute a "set difference" (conceptually) between the list of
+        filtered samples and ones that are already in the NMDC database, i.e., we ignore biosamples
+        that are already present in the database, and continue on to create biosample_set records for
+        those that do not have records in the database already.
+
+        :return: An instance of `nmdc:Database` object which is JSON-ified and rendered on the frontend.
+        """
+        database = nmdc.Database()
+
+        # get a list of all biosamples associated with a given NMDC study id
+        biosample_set = self.runtime_api_user_client.get_biosamples_for_study(
+            self.study_id
+        )
+
+        # get a list of GOLD biosample ids (`biosampleGoldId` values) by iterating
+        # over all the biosample_set records retrieved using the above logic
+        nmdc_gold_ids = set()
+        for biosample in biosample_set:
+            gold_ids = biosample.get("gold_biosample_identifiers", [])
+            for gold_id in gold_ids:
+                nmdc_gold_ids.add(gold_id.replace("gold:", ""))
+
+        # retrieve GOLD study id by looking at the `gold_study_identifiers` key/slot
+        # on the NMDC study record
+        nmdc_study = self.runtime_api_user_client.get_study(self.study_id)[0]
+        gold_study_id = nmdc_study.get("gold_study_identifiers", [])[0].replace(
+            "gold:", ""
+        )
+
+        # use the GOLD study id to fetch all biosample records associated with the study
+        gold_biosamples_for_study = self.gold_api_client.fetch_biosamples_by_study(
+            gold_study_id
+        )
+
+        # part of the code where we are (conceptually) computing a set difference between
+        # the list of filtered samples and ones that are already in the NMDC database
+        missing_gold_biosamples = [
+            gbs
+            for gbs in gold_biosamples_for_study
+            if gbs.get("biosampleGoldId") not in nmdc_gold_ids
+        ]
+
+        gold_study_translator = GoldStudyTranslator(
+            biosamples=missing_gold_biosamples,
+            gold_nmdc_instrument_map_df=self.gold_nmdc_instrument_map_df,
+        )
+
+        translated_biosamples = gold_study_translator.biosamples
+
+        # mint new NMDC biosample IDs for the "missing" biosamples
+        gold_biosample_ids = [
+            biosample["biosampleGoldId"] for biosample in translated_biosamples
+        ]
+        nmdc_biosample_ids = self.runtime_api_site_client.mint_id(
+            "nmdc:Biosample", len(translated_biosamples)
+        ).json()
+        gold_to_nmdc_biosample_ids = dict(zip(gold_biosample_ids, nmdc_biosample_ids))
+
+        database.biosample_set = [
+            gold_study_translator._translate_biosample(
+                biosample,
+                nmdc_biosample_id=gold_to_nmdc_biosample_ids[
+                    biosample["biosampleGoldId"]
+                ],
+                nmdc_study_id=self.study_id,
+                nmdc_field_site_id=None,
+            )
+            for biosample in translated_biosamples
+        ]
+
+        return database

nmdc_runtime/site/repository.py

@@ -25,6 +25,7 @@ from nmdc_runtime.api.models.run import _add_run_fail_event
 from nmdc_runtime.api.models.trigger import Trigger
 from nmdc_runtime.site.export.study_metadata import export_study_biosamples_metadata
 from nmdc_runtime.site.graphs import (
+    generate_biosample_set_from_samples_in_gold,
     translate_metadata_submission_to_nmdc_schema_database,
     ingest_metadata_submission,
     gold_study_to_database,
@@ -44,6 +45,7 @@ from nmdc_runtime.site.graphs import (
     ingest_neon_surface_water_metadata,
     ensure_alldocs,
     nmdc_study_to_ncbi_submission_export,
+    generate_data_generation_set_for_biosamples_in_nmdc_study,
 )
 from nmdc_runtime.site.resources import (
     get_mongo,
@@ -113,6 +115,13 @@ housekeeping_weekly = ScheduleDefinition(
     job=housekeeping.to_job(**preset_normal),
 )
 
+ensure_alldocs_daily = ScheduleDefinition(
+    name="daily_ensure_alldocs",
+    cron_schedule="0 3 * * *",
+    execution_timezone="America/New_York",
+    job=ensure_alldocs.to_job(**preset_normal),
+)
+
 
 def asset_materialization_metadata(asset_event, key):
     """Get metadata from an asset materialization event.
@@ -453,7 +462,7 @@ def repo():
         export_study_biosamples_metadata.to_job(**preset_normal),
         ensure_alldocs.to_job(**preset_normal),
     ]
-    schedules = [housekeeping_weekly]
+    schedules = [housekeeping_weekly, ensure_alldocs_daily]
     sensors = [
         done_object_put_ops,
         ensure_gold_translation_job,
@@ -643,7 +652,7 @@ def biosample_submission_ingest():
                         "inputs": {
                             "neon_envo_mappings_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/main/assets/neon_mixs_env_triad_mappings/neon-nlcd-local-broad-mappings.tsv",
                             "neon_raw_data_file_mappings_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/main/assets/misc/neon_raw_data_file_mappings.tsv",
-                            "neon_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/berkeley-schema-fy24/refs/heads/main/assets/misc/neon_sequencingMethod_to_nmdc_instrument_set.tsv",
+                            "neon_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/refs/heads/main/assets/misc/neon_sequencingMethod_to_nmdc_instrument_set.tsv",
                         }
                     },
                 },
@@ -685,7 +694,7 @@ def biosample_submission_ingest():
                         "inputs": {
                             "neon_envo_mappings_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/main/assets/neon_mixs_env_triad_mappings/neon-nlcd-local-broad-mappings.tsv",
                             "neon_raw_data_file_mappings_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/main/assets/misc/neon_raw_data_file_mappings.tsv",
-                            "neon_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/berkeley-schema-fy24/refs/heads/main/assets/misc/neon_sequencingMethod_to_nmdc_instrument_set.tsv",
+                            "neon_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/refs/heads/main/assets/misc/neon_sequencingMethod_to_nmdc_instrument_set.tsv",
                         }
                     },
                 },
@@ -728,7 +737,7 @@ def biosample_submission_ingest():
                         "inputs": {
                             "neon_envo_mappings_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/main/assets/neon_mixs_env_triad_mappings/neon-nlcd-local-broad-mappings.tsv",
                             "neon_raw_data_file_mappings_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/main/assets/misc/neon_raw_data_file_mappings.tsv",
-                            "neon_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/berkeley-schema-fy24/refs/heads/main/assets/misc/neon_sequencingMethod_to_nmdc_instrument_set.tsv",
+                            "neon_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/refs/heads/main/assets/misc/neon_sequencingMethod_to_nmdc_instrument_set.tsv",
                         }
                     },
                     "get_neon_pipeline_benthic_data_product": {
@@ -770,7 +779,7 @@ def biosample_submission_ingest():
                         "inputs": {
                             "neon_envo_mappings_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/main/assets/neon_mixs_env_triad_mappings/neon-nlcd-local-broad-mappings.tsv",
                             "neon_raw_data_file_mappings_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/main/assets/misc/neon_raw_data_file_mappings.tsv",
-                            "neon_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/berkeley-schema-fy24/refs/heads/main/assets/misc/neon_sequencingMethod_to_nmdc_instrument_set.tsv",
+                            "neon_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/refs/heads/main/assets/misc/neon_sequencingMethod_to_nmdc_instrument_set.tsv",
                         }
                     },
                 },
@@ -813,14 +822,14 @@ def biosample_submission_ingest():
                         "inputs": {
                             "neon_envo_mappings_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/main/assets/neon_mixs_env_triad_mappings/neon-nlcd-local-broad-mappings.tsv",
                             "neon_raw_data_file_mappings_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/main/assets/misc/neon_raw_data_file_mappings.tsv",
-                            "neon_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/berkeley-schema-fy24/refs/heads/main/assets/misc/neon_sequencingMethod_to_nmdc_instrument_set.tsv",
+                            "neon_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/refs/heads/main/assets/misc/neon_sequencingMethod_to_nmdc_instrument_set.tsv",
                         }
                     },
                     "get_neon_pipeline_surface_water_data_product": {
                         "config": {
                             "surface_water_data_product": {
                                 "product_id": "DP1.20281.001",
-                                "product_tables": "mms_swMetagenomeSequencing, mms_swMetagenomeDnaExtraction, amc_fieldGenetic, amc_fieldSuperParent",
+                                "product_tables": "mms_swMetagenomeSequencing, mms_swMetagenomeDnaExtraction, amc_fieldGenetic, amc_fieldSuperParent, mms_swRawDataFiles",
                             }
                         }
                     },
@@ -847,7 +856,7 @@ def biosample_submission_ingest():
                         "config": {
                             "surface_water_data_product": {
                                 "product_id": "DP1.20281.001",
-                                "product_tables": "mms_swMetagenomeSequencing, mms_swMetagenomeDnaExtraction, amc_fieldGenetic, amc_fieldSuperParent",
+                                "product_tables": "mms_swMetagenomeSequencing, mms_swMetagenomeDnaExtraction, amc_fieldGenetic, amc_fieldSuperParent, mms_swRawDataFiles",
                             }
                         }
                     },
@@ -855,7 +864,7 @@ def biosample_submission_ingest():
                         "inputs": {
                             "neon_envo_mappings_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/main/assets/neon_mixs_env_triad_mappings/neon-nlcd-local-broad-mappings.tsv",
                             "neon_raw_data_file_mappings_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/main/assets/misc/neon_raw_data_file_mappings.tsv",
-                            "neon_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/berkeley-schema-fy24/refs/heads/main/assets/misc/neon_sequencingMethod_to_nmdc_instrument_set.tsv",
+                            "neon_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/refs/heads/main/assets/misc/neon_sequencingMethod_to_nmdc_instrument_set.tsv",
                         }
                     },
                 },
@@ -915,6 +924,97 @@ def biosample_export():
     ]
 
 
+@repository
+def database_records_stitching():
+    normal_resources = run_config_frozen__normal_env["resources"]
+    return [
+        generate_data_generation_set_for_biosamples_in_nmdc_study.to_job(
+            description="This job can be used to create a data_generation_set JSON for biosamples that are already present in the NMDC database.",
+            resource_defs=resource_defs,
+            config={
+                "resources": merge(
+                    unfreeze(normal_resources),
+                    {
+                        "runtime_api_user_client": {
+                            "config": {
+                                "base_url": {"env": "API_HOST"},
+                                "username": {"env": "API_ADMIN_USER"},
+                                "password": {"env": "API_ADMIN_PASS"},
+                            },
+                        },
+                        "runtime_api_site_client": {
+                            "config": {
+                                "base_url": {"env": "API_HOST"},
+                                "client_id": {"env": "API_SITE_CLIENT_ID"},
+                                "client_secret": {"env": "API_SITE_CLIENT_SECRET"},
+                                "site_id": {"env": "API_SITE_ID"},
+                            },
+                        },
+                        "gold_api_client": {
+                            "config": {
+                                "base_url": {"env": "GOLD_API_BASE_URL"},
+                                "username": {"env": "GOLD_API_USERNAME"},
+                                "password": {"env": "GOLD_API_PASSWORD"},
+                            },
+                        },
+                    },
+                ),
+                "ops": {
+                    "get_database_updater_inputs": {
+                        "config": {
+                            "nmdc_study_id": "",
+                            "gold_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/refs/heads/main/assets/misc/gold_seqMethod_to_nmdc_instrument_set.tsv",
+                        }
+                    },
+                    "export_json_to_drs": {"config": {"username": ""}},
+                },
+            },
+        ),
+        generate_biosample_set_from_samples_in_gold.to_job(
+            description="This job can be used to create a biosample_set JSON from samples in GOLD for a given study in NMDC.",
+            resource_defs=resource_defs,
+            config={
+                "resources": merge(
+                    unfreeze(normal_resources),
+                    {
+                        "runtime_api_user_client": {
+                            "config": {
+                                "base_url": {"env": "API_HOST"},
+                                "username": {"env": "API_ADMIN_USER"},
+                                "password": {"env": "API_ADMIN_PASS"},
+                            },
+                        },
+                        "runtime_api_site_client": {
+                            "config": {
+                                "base_url": {"env": "API_HOST"},
+                                "client_id": {"env": "API_SITE_CLIENT_ID"},
+                                "client_secret": {"env": "API_SITE_CLIENT_SECRET"},
+                                "site_id": {"env": "API_SITE_ID"},
+                            },
+                        },
+                        "gold_api_client": {
+                            "config": {
+                                "base_url": {"env": "GOLD_API_BASE_URL"},
+                                "username": {"env": "GOLD_API_USERNAME"},
+                                "password": {"env": "GOLD_API_PASSWORD"},
+                            },
+                        },
+                    },
+                ),
+                "ops": {
+                    "get_database_updater_inputs": {
+                        "config": {
+                            "nmdc_study_id": "",
+                            "gold_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/refs/heads/main/assets/misc/gold_seqMethod_to_nmdc_instrument_set.tsv",
+                        }
+                    },
+                    "export_json_to_drs": {"config": {"username": ""}},
+                },
+            },
+        ),
+    ]
+
+
 # @repository
 # def validation():
 #     graph_jobs = [validate_jgi_job, validate_gold_job, validate_emsl_job]

nmdc_runtime/site/resources.py

@@ -129,16 +129,23 @@ class RuntimeApiUserClient(RuntimeApiClient):
         return response.json()["cursor"]["firstBatch"]
 
     def get_biosamples_for_study(self, study_id: str):
+        # TODO: 10000 is an arbitrarily large number that has been chosen for the max_page_size param.
+        # The /nmdcschema/{collection-name} endpoint implements pagination via the page_token mechanism,
+        # but the tradeoff there is that we would need to make multiple requests to step through the
+        # each of the pages. By picking a large number for max_page_size, we can get all the results
+        # in a single request.
+        # This method previously used the /queries:run endpoint but the problem with that was that
+        # it used to truncate the number of results returned to 100.
         response = self.request(
-            "POST",
-            f"/queries:run",
+            "GET",
+            f"/nmdcschema/biosample_set",
             {
-                "find": "biosample_set",
-                "filter": {"part_of": {"$elemMatch": {"$eq": study_id}}},
+                "filter": json.dumps({"associated_studies": study_id}),
+                "max_page_size": 10000,
             },
         )
         response.raise_for_status()
-        return response.json()["cursor"]["firstBatch"]
+        return response.json()["resources"]
 
     def get_omics_processing_by_name(self, name: str):
         response = self.request(
@@ -152,6 +159,18 @@ class RuntimeApiUserClient(RuntimeApiClient):
         response.raise_for_status()
         return response.json()["cursor"]["firstBatch"]
 
+    def get_study(self, study_id: str):
+        response = self.request(
+            "POST",
+            f"/queries:run",
+            {
+                "find": "study_set",
+                "filter": {"id": study_id},
+            },
+        )
+        response.raise_for_status()
+        return response.json()["cursor"]["firstBatch"]
+
 
 class RuntimeApiSiteClient(RuntimeApiClient):
     def __init__(
@@ -370,6 +389,18 @@ class GoldApiClient(BasicAuthClient):
             return None
         return results[0]
 
+    def fetch_projects_by_biosample(self, biosample_id: str) -> List[Dict[str, Any]]:
+        id = self._normalize_id(biosample_id)
+        results = self.request("/projects", params={"biosampleGoldId": id})
+        return results
+
+    def fetch_biosample_by_biosample_id(
+        self, biosample_id: str
+    ) -> List[Dict[str, Any]]:
+        id = self._normalize_id(biosample_id)
+        results = self.request("/biosamples", params={"biosampleGoldId": id})
+        return results
+
 
 @resource(
     config_schema={

nmdc_runtime/site/translation/gold_translator.py

@@ -12,6 +12,29 @@ from nmdc_runtime.site.translation.translator import JSON_OBJECT, Translator
 SEQUENCING_STRATEGIES = {"Metagenome", "Metatranscriptome"}
 
 
+def _is_valid_project(project: dict) -> bool:
+    """A project is considered valid if:
+    1. `sequencingStrategy` is in {"Metagenome", "Metatranscriptome"}
+    2. if `sequencingCenters` == 'DOE Joint Genome Institute (JGI)' then
+        `projectStatus` must be in ("Permanent Draft", "Complete and Published")
+    3. otherwise, no `projectStatus` filter is applied
+
+    :param project: GOLD project object (structurally similar to response
+                    from `/projects` endpoint)
+    :return: True if the project is valid, False otherwise
+    """
+    if project.get("sequencingStrategy") not in SEQUENCING_STRATEGIES:
+        return False
+
+    if project.get("sequencingCenters") == "DOE Joint Genome Institute (JGI)":
+        return project.get("projectStatus") in (
+            "Permanent Draft",
+            "Complete and Published",
+        )
+
+    return True
+
+
 class GoldStudyTranslator(Translator):
     def __init__(
         self,
@@ -36,16 +59,15 @@ class GoldStudyTranslator(Translator):
             biosample
             for biosample in biosamples
             if any(
-                project.get("sequencingStrategy") in SEQUENCING_STRATEGIES
-                for project in biosample.get("projects", [])
+                _is_valid_project(project) for project in biosample.get("projects", [])
             )
         ]
         # Fetch the valid projectGoldIds that are associated with filtered
         # biosamples on their `projects` field
         valid_project_ids = {
             project.get("projectGoldId")
-            for biosample in self.biosamples
-            for project in biosample.get("projects", [])
+            for project in projects
+            if _is_valid_project(project)
         }
         # Filter projects to only those with `projectGoldId` in valid_project_ids
         self.projects = [