Make the choice of "accession" on the config level

There was useful information on why "accession" is used over "strain" in
the docstring of update_example_data. Update that rule to use the
strain_id_field config, and move the context to every config file that
sets strain_id_field as "accession".

Don't set defaults when retrieving strain_id_field so "accession" is
only set on the config level.

config/config_hmpxv1.yaml

@@ -9,6 +9,8 @@ auspice_config: "config/auspice_config_hmpxv1.json"
 description: "config/description.md"
 tree_mask: "config/tree_mask.tsv"
 
+# Use `accession` as the ID column since `strain` currently contains duplicates¹.
+# ¹ https://github.com/nextstrain/monkeypox/issues/33
 strain_id_field: "accession"
 display_strain_field: "strain"
 

config/config_hmpxv1_big.yaml

@@ -9,6 +9,8 @@ auspice_config: "config/auspice_config_hmpxv1_big.json"
 description: "config/description.md"
 tree_mask: "config/tree_mask.tsv"
 
+# Use `accession` as the ID column since `strain` currently contains duplicates¹.
+# ¹ https://github.com/nextstrain/monkeypox/issues/33
 strain_id_field: "accession"
 display_strain_field: "strain"
 

config/config_mpxv.yaml

@@ -9,6 +9,8 @@ description: "config/description.md"
 clades: "config/clades.tsv"
 tree_mask: "config/tree_mask.tsv"
 
+# Use `accession` as the ID column since `strain` currently contains duplicates¹.
+# ¹ https://github.com/nextstrain/monkeypox/issues/33
 strain_id_field: "accession"
 display_strain_field: "strain"
 

workflow/snakemake_rules/chores.smk

@@ -5,9 +5,6 @@ rule update_example_data:
     - sets the subsampling size to 50
     - includes the root (defined in config but hardcoded here)
     - ensures all clades and lineages are accounted for using --group-by
-    - uses `accession` as the ID column since `strain` currently contains duplicates
-
-    TODO: Use `strain` as the ID column after https://github.com/nextstrain/monkeypox/issues/33 is done.
     """
     message:
         "Update example data"
@@ -17,11 +14,13 @@ rule update_example_data:
     output:
         sequences="example_data/sequences.fasta",
         metadata="example_data/metadata.tsv",
+    params:
+        strain_id=lambda w: config.get("strain_id_field"),
     shell:
         """
         augur filter \
             --metadata {input.metadata} \
-            --metadata-id-columns accession \
+            --metadata-id-columns {params.strain_id} \
             --sequences {input.sequences} \
             --include-where strain=MK783032 strain=MK783030 \
             --group-by clade lineage \

workflow/snakemake_rules/core.smk

@@ -32,7 +32,7 @@ rule exclude_bad:
     params:
         min_date=config["min_date"],
         min_length=config["min_length"],
-        strain_id=lambda w: config.get("strain_id_field", "accession"),
+        strain_id=lambda w: config.get("strain_id_field"),
     shell:
         """
         augur filter \
@@ -81,7 +81,7 @@ rule filter:
         group_by=config.get("group_by", "--group-by clade lineage"),
         sequences_per_group=config["sequences_per_group"],
         other_filters=config.get("filters", ""),
-        strain_id=lambda w: config.get("strain_id_field", "accession"),
+        strain_id=lambda w: config.get("strain_id_field"),
     shell:
         """
         augur filter \
@@ -236,7 +236,7 @@ rule refine:
         clock_std_dev=lambda w: f"--clock-std-dev {config['clock_std_dev']}"
         if "clock_std_dev" in config
         else "",
-        strain_id=lambda w: config.get("strain_id_field", "accession"),
+        strain_id=lambda w: config.get("strain_id_field"),
     shell:
         """
         augur refine \
@@ -312,7 +312,7 @@ rule traits:
     params:
         columns="country",
         sampling_bias_correction=3,
-        strain_id=lambda w: config.get("strain_id_field", "accession"),
+        strain_id=lambda w: config.get("strain_id_field"),
     shell:
         """
         augur traits \
@@ -444,7 +444,7 @@ rule export:
         auspice_json=build_dir + "/{build_name}/raw_tree.json",
         root_sequence=build_dir + "/{build_name}/raw_tree_root-sequence.json",
     params:
-        strain_id=lambda w: config.get("strain_id_field", "accession"),
+        strain_id=lambda w: config.get("strain_id_field"),
     shell:
         """
         augur export v2 \