text using cchf

loculus-project · Jan 10, 2025 · a41e77f · a41e77f
1 parent 87a7f6a
commit a41e77f
Show file tree

Hide file tree

Showing 2 changed files with 13 additions and 8 deletions.
diff --git a/ingest/scripts/deterministic_group_segments.py b/ingest/scripts/deterministic_group_segments.py
@@ -114,7 +114,7 @@ def group_records(record_list, output_metadata, fasta_id_map, config, different_
 
     joint_key = "/".join(
         [
-            f"{segment_map[segment]}.{segment}"
+            f"{segment_map[segment]['insdcAccessionFull']}.{segment}"
             for segment in config.nucleotide_sequences
             if segment in segment_map
         ]
@@ -177,12 +177,13 @@ def main(
     logger.info(f"Reading metadata from {input_metadata}")
 
     source_of_truth = json.load(open(groups, encoding="utf-8"))
+    logger.info(f"Found {len(source_of_truth.keys())} source of truth groups")
     accession_to_group = {}
     for group, metadata in source_of_truth.items():
         for accession in metadata:
             accession_to_group[accession] = group
 
-    found_groups = {}
+    found_groups = {group: [] for group in source_of_truth}
     # Map from original accession to the new concatenated accession
     type Accession = str
     type SubmissionId = str
@@ -202,10 +203,7 @@ def main(
             ungrouped_accessions.add(record["id"])
             continue
         group = accession_to_group[metadata["insdcAccessionFull"]]
-        if group not in found_groups:
-            found_groups[group] = [record]
-        else:
-            found_groups[group].append(record)
+        found_groups[group].append(record)
         if len(found_groups[group]) == len(set(source_of_truth[group])):
             group_records(
                 found_groups[group], output_metadata, fasta_id_map, config, different_values_log
@@ -217,13 +215,19 @@ def main(
 
     # add found_groups without all segments in file
     count_unfilled_groups = 0
+    count_missing_tests = 0
     for name, records in found_groups.items():
         count_unfilled_groups += 1
-        logger.debug(f"{name}: Missing record {set(source_of_truth[name]) - set([record['metadata']['insdcAccessionFull'] for record in records])}")
+        logger.debug(
+            f"{name}: Missing record {set(source_of_truth[name]) - {record['metadata']['insdcAccessionFull'] for record in records}}"
+        )
+        if len(records) == 0:
+            count_missing_tests += 1
+            continue
         group_records(records, output_metadata, fasta_id_map, config, different_values_log)
     logger.info(different_values_log)
     logger.info(f"Found {count_unfilled_groups} groups without all segments")
-
+    logger.info(f"Found {count_missing_tests} groups without any segments")
 
     count = 0
     count_ignored = 0

diff --git a/kubernetes/loculus/values.yaml b/kubernetes/loculus/values.yaml
@@ -1487,6 +1487,7 @@ defaultOrganisms:
         taxon_id: 3052518
         nextclade_dataset_server: https://raw.githubusercontent.com/nextstrain/nextclade_data/cornelius-cchfv/data_output
         nextclade_dataset_name: nextstrain/cchfv/linked
+        grouping_ground_truth: https://anna-parker.github.io/influenza-a-groupings/results/cchf_groups.json
     enaDeposition:
       configFile:
         taxon_id: 3052518