ID strain by accession

nextstrain · Sep 13, 2023 · 65d73a7 · 65d73a7
1 parent 81a072a
commit 65d73a7
Show file tree

Hide file tree

Showing 2 changed files with 19 additions and 5 deletions.
diff --git a/Snakefile b/Snakefile
@@ -1,3 +1,6 @@
+if not config:
+    configfile: "config/config_dengue.yaml"
+
 serotypes = ['all', 'denv1', 'denv2', 'denv3', 'denv4']
 
 rule all:
@@ -98,12 +101,14 @@ rule filter:
     params:
         group_by = "year region",
         sequences_per_group = filter_sequences_per_group,
-        min_length = 5000
+        min_length = 5000,
+        strain_id = config.get("strain_id_field", "strain"),
     shell:
         """
         augur filter \
             --sequences {input.sequences} \
             --metadata {input.metadata} \
+            --metadata-id-columns {params.strain_id} \
             --exclude {input.exclude} \
             --output {output.sequences} \
             --group-by {params.group_by} \
@@ -161,17 +166,19 @@ rule refine:
         metadata = "results/metadata_{serotype}.tsv"
     output:
         tree = "results/tree_{serotype}.nwk",
-        node_data = "results/branch-lengths_{serotype}.json"
+        node_data = "results/branch-lengths_{serotype}.json",
     params:
         coalescent = "const",
         date_inference = "marginal",
-        clock_filter_iqd = 4
+        clock_filter_iqd = 4,
+        strain_id = config.get("strain_id_field", "strain"),
     shell:
         """
         augur refine \
             --tree {input.tree} \
             --alignment {input.alignment} \
             --metadata {input.metadata} \
+            --metadata-id-columns {params.strain_id} \
             --output-tree {output.tree} \
             --output-node-data {output.node_data} \
             --timetree \
@@ -228,12 +235,14 @@ rule traits:
         node_data = "results/traits_{serotype}.json",
     params:
         columns = traits_columns,
-        sampling_bias_correction = 3
+        sampling_bias_correction = 3,
+        strain_id = config.get("strain_id_field", "strain"),
     shell:
         """
         augur traits \
             --tree {input.tree} \
             --metadata {input.metadata} \
+            --metadata-id-columns {params.strain_id} \
             --output {output.node_data} \
             --columns {params.columns} \
             --confidence \
@@ -270,12 +279,15 @@ rule export:
         aa_muts = "results/aa-muts_{serotype}.json",
         auspice_config = files.auspice_config
     output:
-        auspice_json = "auspice/dengue_{serotype}.json"
+        auspice_json = "auspice/dengue_{serotype}.json",
+    params:
+        strain_id = config.get("strain_id_field", "strain"),
     shell:
         """
         augur export v2 \
             --tree {input.tree} \
             --metadata {input.metadata} \
+            --metadata-id-columns {params.strain_id} \
             --node-data {input.branch_lengths} {input.traits} {input.clades} {input.nt_muts} {input.aa_muts} \
             --auspice-config {input.auspice_config} \
             --include-root-sequence \

diff --git a/config/config_dengue.yaml b/config/config_dengue.yaml
@@ -0,0 +1,2 @@
+strain_id_field: "accession"
+display_strain_field: "strain"