phylogenetic: Move input URLs to config YAML

Allows us to easily update the input URLs both from the config YAML and `--config` args. This is mostly motivated by the desire to run phylogenetic builds from trial ingest outputs.
nextstrain · Apr 12, 2024 · 27c6557 · 27c6557
1 parent fa3ba47
commit 27c6557
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 3 deletions.
diff --git a/phylogenetic/defaults/config_zika.yaml b/phylogenetic/defaults/config_zika.yaml
@@ -1,3 +1,8 @@
+# Sequences must be FASTA and metadata must be TSV
+# Both files must be zstd compressed
+sequences_url: "https://data.nextstrain.org/files/workflows/zika/sequences.fasta.zst"
+metadata_url: "https://data.nextstrain.org/files/workflows/zika/metadata.tsv.zst"
+
 strain_id_field: "accession"
 display_strain_field: "strain"
 

diff --git a/phylogenetic/rules/prepare_sequences.smk b/phylogenetic/rules/prepare_sequences.smk
@@ -27,8 +27,8 @@ rule download:
         sequences = "data/sequences.fasta.zst",
         metadata = "data/metadata.tsv.zst"
     params:
-        sequences_url = "https://data.nextstrain.org/files/workflows/zika/sequences.fasta.zst",
-        metadata_url = "https://data.nextstrain.org/files/workflows/zika/metadata.tsv.zst"
+        sequences_url = config["sequences_url"],
+        metadata_url = config["metadata_url"],
     shell:
         """
         curl -fsSL --compressed {params.sequences_url:q} --output {output.sequences}
@@ -101,4 +101,4 @@ rule align:
             --output {output.alignment} \
             --fill-gaps \
             --remove-reference
-        """
+        """