agg does not work with strings :/

broadinstitute · Nov 28, 2024 · c070c7c · c070c7c
1 parent cf242f1
commit c070c7c
Showing 1 changed file with 4 additions and 1 deletion.
diff --git a/v03_pipeline/lib/reference_datasets/dbnsfp.py b/v03_pipeline/lib/reference_datasets/dbnsfp.py
@@ -79,7 +79,10 @@ def get_ht(path: str, reference_genome: ReferenceGenome) -> hl.Table:
         ht = ht.rename(rename)
         ht = key_by_locus_alleles(ht, reference_genome)
         return ht.group_by(*ht.key).aggregate(
-            **{f: hl.agg.max(ht[f]) for f in ht.row_value},
+            MutationTaster_pred=hl.agg.take(ht.MutationTaster_pred, 1)[0],
+            **{
+                f: hl.agg.max(ht[f]) for f in ht.row_value if f != 'MutationTaster_pred'
+            },
         )