diff --git a/ingest/workflow/snakemake_rules/split_serotypes.smk b/ingest/workflow/snakemake_rules/split_serotypes.smk
index b389fa8a..6ac75767 100644
--- a/ingest/workflow/snakemake_rules/split_serotypes.smk
+++ b/ingest/workflow/snakemake_rules/split_serotypes.smk
@@ -14,30 +14,30 @@ Parameters are expected to be defined in `config.transform`.
 """
 
 rule split_by_ncbi_serotype:
-  """
-  Split the data by serotype based on the NCBI metadata.
-  """
-  input:
-    metadata = "data/metadata.tsv",
-    sequences = "data/sequences.fasta"
-  output:
-    metadata = "results/metadata_{serotype}.tsv",
-    sequences = "results/sequences_{serotype}.fasta"
-  params:
-    ncbi_serotype = lambda wildcards: wildcards.serotype,
-    id_field = config["transform"]["id_field"]
-  shell:
     """
-    if [[ "{params.ncbi_serotype}" == "all" ]]; then
-      cp {input.metadata} {output.metadata}
-      cp {input.sequences} {output.sequences}
-    else
-      augur filter \
-        --sequences {input.sequences} \
-        --metadata {input.metadata} \
-        --metadata-id-columns {params.id_field} \
-        --query "ncbi_serotype=='{params.ncbi_serotype}'" \
-        --output-sequences {output.sequences} \
-        --output-metadata {output.metadata}
-    fi
+    Split the data by serotype based on the NCBI metadata.
     """
+    input:
+        metadata = "data/metadata.tsv",
+        sequences = "data/sequences.fasta"
+    output:
+        metadata = "results/metadata_{serotype}.tsv",
+        sequences = "results/sequences_{serotype}.fasta"
+    params:
+        ncbi_serotype = lambda wildcards: wildcards.serotype,
+        id_field = config["transform"]["id_field"]
+    shell:
+        """
+        if [[ "{params.ncbi_serotype}" == "all" ]]; then
+          cp {input.metadata} {output.metadata}
+          cp {input.sequences} {output.sequences}
+        else
+          augur filter \
+            --sequences {input.sequences} \
+            --metadata {input.metadata} \
+            --metadata-id-columns {params.id_field} \
+            --query "ncbi_serotype=='{params.ncbi_serotype}'" \
+            --output-sequences {output.sequences} \
+            --output-metadata {output.metadata}
+        fi
+        """