Merge branch 'refs/heads/master' into release_3

hubmapconsortium · Oct 11, 2024 · d0ffb4b · d0ffb4b
2 parents 7928d74 + 791181a
commit d0ffb4b
Show file tree

Hide file tree

Showing 2 changed files with 15 additions and 11 deletions.
diff --git a/src/ingest-pipeline/airflow/dags/azimuth_annotations.py b/src/ingest-pipeline/airflow/dags/azimuth_annotations.py
@@ -88,15 +88,15 @@ def build_cwltool_cmd1(**kwargs):
 
         organ_list = list(set(ds_rslt["organs"]))
         organ_code = organ_list[0] if len(organ_list) == 1 else "multi"
-        assay = get_assay_previous_version(**kwargs)
+        assay, matrix = get_assay_previous_version(**kwargs)
 
         command = [
             *get_cwltool_base_cmd(tmpdir),
             cwl_workflows_annotations[0],
             "--reference",
             organ_code,
             "--matrix",
-            "expr.h5ad",
+            matrix,
             "--secondary-analysis-matrix",
             "secondary_analysis.h5ad",
             "--assay",

diff --git a/src/ingest-pipeline/airflow/dags/utils.py b/src/ingest-pipeline/airflow/dags/utils.py
@@ -351,22 +351,26 @@ def my_callable(**kwargs):
     return ds_rslt["dataset_info"]
 
 
-def get_assay_previous_version(**kwargs) -> str:
+def get_assay_previous_version(**kwargs) -> tuple:
     dataset_type = get_dataname_previous_version(**kwargs).split("__")[0]
     if dataset_type == "salmon_rnaseq_10x":
-        return "10x_v3"
+        return "10x_v3", "expr.h5ad"
     if dataset_type == "salmon_rnaseq_10x_sn":
-        return "10x_v3_sn"
+        return "10x_v3_sn", "expr.h5ad"
     if dataset_type == "salmon_rnaseq_10x_v2":
-        return "10x_v2"
+        return "10x_v2", "expr.h5ad"
     if dataset_type == "salmon_rnaseq_10x_v2_sn":
-        return "10x_v2_sn"
+        return "10x_v2_sn", "expr.h5ad"
     if dataset_type == "salmon_rnaseq_sciseq":
-        return "sciseq"
-    if dataset_type == "salmon_rnaseq_snareseq" or dataset_type == "multiome_snareseq":
-        return "snareseq"
+        return "sciseq", "expr.h5ad"
+    if dataset_type == "salmon_rnaseq_snareseq":
+        return "snareseq", "expr.h5ad"
     if dataset_type == "salmon_rnaseq_slideseq":
-        return "slideseq"
+        return "slideseq", "expr.h5ad"
+    if dataset_type == "multiome_10x":
+        return "10x_V3_sn", "mudata_raw.h5mu"
+    if dataset_type == "multiome_snareseq":
+        return "snareseq", "mudata_raw.h5mu"
 
 
 def get_parent_dataset_paths_list(**kwargs) -> List[Path]: