rules.diff

diff --git a/ingest/defaults/config.yaml b/ingest/defaults/config.yaml
index 7bd0362..dfddd22 100644
--- a/ingest/defaults/config.yaml
+++ b/ingest/defaults/config.yaml
@@ -39,9 +39,11 @@ curate:
   # For the Nextstrain team, this is currently
   # "https://raw.githubusercontent.com/nextstrain/ncov-ingest/master/source-data/gisaid_geoLocationRules.tsv"
   geolocation_rules_url: "https://raw.githubusercontent.com/nextstrain/ncov-ingest/master/source-data/gisaid_geoLocationRules.tsv"
-  # The path to the local geolocation rules within the pathogen repo
+  # General global defaults are sourced from the geolocation_rules_url URL above.
+  # Pathogen-specific defaults are sourced from defaults/geolocation_rules.tsv.
+  # Local customizations are sourced from the path named here.
   # The path should be relative to the ingest directory.
-  local_geolocation_rules: "defaults/geolocation_rules.tsv"
+  local_geolocation_rules: ~
   # List of field names to change where the key is the original field name and the value is the new field name
   # The original field names should match the ncbi_datasets_fields provided above.
   # This is the first step in the pipeline, so any references to field names in the configs below should use the new field names
@@ -93,10 +95,12 @@ curate:
   authors_default_value: "?"
   # Name to use for the generated abbreviated authors field
   abbr_authors_field: "abbr_authors"
-  # Path to the manual annotations file
-  # The path should be relative to the ingest directory
-  annotations: "defaults/annotations.tsv"
+  # Pathogen-specific default annotations are sourced from defaults/annotations.tsv.
+  # Local customizations are sourced from the path named here.
+  # The path should be relative to the ingest directory.
+  annotations: ~
   # The ID field in the metadata to use to merge the manual annotations
+  # XXX FIXME: address differing id fields between defaults and local
   annotations_id: "accession"
   # The ID field in the metadata to use as the sequence id in the output FASTA file
   output_id_field: "accession"
diff --git a/ingest/rules/curate.smk b/ingest/rules/curate.smk
index 71ffa18..bfa088e 100644
--- a/ingest/rules/curate.smk
+++ b/ingest/rules/curate.smk
@@ -13,10 +13,11 @@ OUTPUTS:
 """
 
 
-# The following two rules can be ignored if you choose not to use the
+# The following rule can be ignored if you choose not to use the
 # generalized geolocation rules that are shared across pathogens.
 # The Nextstrain team will try to maintain a generalized set of geolocation
-# rules that can then be overridden by local geolocation rules per pathogen repo.
+# rules that can then be overridden by additional geolocation rules per
+# pathogen repo.
 rule fetch_general_geolocation_rules:
     output:
         general_geolocation_rules="data/general-geolocation-rules.tsv",
@@ -30,13 +31,28 @@ rule fetch_general_geolocation_rules:
 
 rule concat_geolocation_rules:
     input:
-        general_geolocation_rules="data/general-geolocation-rules.tsv",
-        local_geolocation_rules=config["curate"]["local_geolocation_rules"],
-    output:
-        all_geolocation_rules="data/all-geolocation-rules.tsv",
+        *filter(None, [
+            "data/general-geolocation-rules.tsv", # Remove if not using above rule
+            "defaults/geolocation_rules.tsv",
+            config["curate"]["local_geolocation_rules"],
+        ])
+    output: "data/all-geolocation-rules.tsv",
     shell:
         """
-        cat {input.general_geolocation_rules} {input.local_geolocation_rules} >> {output.all_geolocation_rules}
+        cat {input:q} > {output:q}
+        """
+
+
+rule concat_annotations:
+    input:
+        *filter(None, [
+            "defaults/annotations.tsv",
+            config["curate"]["annotations"],
+        ])
+    output: "data/all-annotations.tsv",
+    shell:
+        """
+        cat {input:q} > {output:q}
         """
 
 
@@ -57,9 +73,8 @@ def format_field_map(field_map: dict[str, str]) -> str:
 rule curate:
     input:
         sequences_ndjson="data/ncbi.ndjson",
-        # Change the geolocation_rules input path if you are removing the above two rules
         all_geolocation_rules="data/all-geolocation-rules.tsv",
-        annotations=config["curate"]["annotations"],
+        annotations="data/all-annotations.tsv",
     output:
         metadata="data/all_metadata.tsv",
         sequences="results/sequences.fasta",