fixup! 🚧 Use population-based weighted sampling for Asia builds

nextstrain · Jun 7, 2024 · 88c1397 · 88c1397
1 parent f68c86f
commit 88c1397
Show file tree

Hide file tree

Showing 3 changed files with 15 additions and 14 deletions.
diff --git a/nextstrain_profiles/nextstrain-gisaid/builds.yaml b/nextstrain_profiles/nextstrain-gisaid/builds.yaml
@@ -186,7 +186,7 @@ subsampling:
     # Early focal samples for Asia
     asia_early:
       group_by: "country year month"
-      group_by_weights: "data/country_population_weights.tsv"
+      group_by_weights: "defaults/population_weights.tsv"
       max_sequences: 700
       max_date: "--max-date 1M"
       exclude: "--exclude-where 'region!=Asia'"
@@ -199,7 +199,7 @@ subsampling:
     # Recent focal samples for Asia
     asia_recent:
       group_by: "country year month"
-      group_by_weights: "data/country_population_weights.tsv"
+      group_by_weights: "defaults/population_weights.tsv"
       max_sequences: 2800
       min_date: "--min-date 1M"
       exclude: "--exclude-where 'region!=Asia'"
@@ -220,7 +220,7 @@ subsampling:
     # Early focal samples for Asia
     asia_early:
       group_by: "country year month"
-      group_by_weights: "data/country_population_weights.tsv"
+      group_by_weights: "defaults/population_weights.tsv"
       max_sequences: 700
       max_date: "--max-date 2M"
       exclude: "--exclude-where 'region!=Asia'"
@@ -233,7 +233,7 @@ subsampling:
     # Recent focal samples for Asia
     asia_recent:
       group_by: "country year month"
-      group_by_weights: "data/country_population_weights.tsv"
+      group_by_weights: "defaults/population_weights.tsv"
       max_sequences: 2800
       min_date: "--min-date 2M"
       exclude: "--exclude-where 'region!=Asia'"
@@ -254,7 +254,7 @@ subsampling:
     # Early focal samples for Asia
     asia_early:
       group_by: "country year month"
-      group_by_weights: "data/country_population_weights.tsv"
+      group_by_weights: "defaults/population_weights.tsv"
       max_sequences: 700
       max_date: "--max-date 6M"
       exclude: "--exclude-where 'region!=Asia'"
@@ -267,7 +267,7 @@ subsampling:
     # Recent focal samples for Asia
     asia_recent:
       group_by: "country year month"
-      group_by_weights: "data/country_population_weights.tsv"
+      group_by_weights: "defaults/population_weights.tsv"
       max_sequences: 2800
       min_date: "--min-date 6M"
       exclude: "--exclude-where 'region!=Asia'"
@@ -286,7 +286,7 @@ subsampling:
     # Focal samples for Asia
     asia:
       group_by: "country year month"
-      group_by_weights: "data/country_population_weights.tsv"
+      group_by_weights: "defaults/population_weights.tsv"
       max_sequences: 3500
       exclude: "--exclude-where 'region!=Asia'"
     # Contextual samples from the rest of the world

diff --git a/nextstrain_profiles/nextstrain-open/builds.yaml b/nextstrain_profiles/nextstrain-open/builds.yaml
@@ -186,7 +186,7 @@ subsampling:
     # Early focal samples for Asia
     asia_early:
       group_by: "country year month"
-      group_by_weights: "data/country_population_weights.tsv"
+      group_by_weights: "defaults/population_weights.tsv"
       max_sequences: 700
       max_date: "--max-date 1M"
       exclude: "--exclude-where 'region!=Asia'"
@@ -199,7 +199,7 @@ subsampling:
     # Recent focal samples for Asia
     asia_recent:
       group_by: "country year month"
-      group_by_weights: "data/country_population_weights.tsv"
+      group_by_weights: "defaults/population_weights.tsv"
       max_sequences: 2800
       min_date: "--min-date 1M"
       exclude: "--exclude-where 'region!=Asia'"
@@ -220,7 +220,7 @@ subsampling:
     # Early focal samples for Asia
     asia_early:
       group_by: "country year month"
-      group_by_weights: "data/country_population_weights.tsv"
+      group_by_weights: "defaults/population_weights.tsv"
       max_sequences: 700
       max_date: "--max-date 2M"
       exclude: "--exclude-where 'region!=Asia'"
@@ -233,7 +233,7 @@ subsampling:
     # Recent focal samples for Asia
     asia_recent:
       group_by: "country year month"
-      group_by_weights: "data/country_population_weights.tsv"
+      group_by_weights: "defaults/population_weights.tsv"
       max_sequences: 2800
       min_date: "--min-date 2M"
       exclude: "--exclude-where 'region!=Asia'"
@@ -254,7 +254,7 @@ subsampling:
     # Early focal samples for Asia
     asia_early:
       group_by: "country year month"
-      group_by_weights: "data/country_population_weights.tsv"
+      group_by_weights: "defaults/population_weights.tsv"
       max_sequences: 700
       max_date: "--max-date 6M"
       exclude: "--exclude-where 'region!=Asia'"
@@ -267,7 +267,7 @@ subsampling:
     # Recent focal samples for Asia
     asia_recent:
       group_by: "country year month"
-      group_by_weights: "data/country_population_weights.tsv"
+      group_by_weights: "defaults/population_weights.tsv"
       max_sequences: 2800
       min_date: "--min-date 6M"
       exclude: "--exclude-where 'region!=Asia'"
@@ -286,7 +286,7 @@ subsampling:
     # Focal samples for Asia
     asia:
       group_by: "country year month"
-      group_by_weights: "data/country_population_weights.tsv"
+      group_by_weights: "defaults/population_weights.tsv"
       max_sequences: 3500
       exclude: "--exclude-where 'region!=Asia'"
     # Contextual samples from the rest of the world

diff --git a/workflow/snakemake_rules/main_workflow.smk b/workflow/snakemake_rules/main_workflow.smk
@@ -300,6 +300,7 @@ rule subsample:
         "benchmarks/subsample_{build_name}_{subsample}.txt"
     params:
         group_by = _get_specific_subsampling_setting("group_by", optional=True),
+        # FIXME: pull this from config.files.population_weights?
         group_by_weights = _get_specific_subsampling_setting("group_by_weights", optional=True),
         sequences_per_group = _get_specific_subsampling_setting("seq_per_group", optional=True),
         subsample_max_sequences = _get_specific_subsampling_setting("max_sequences", optional=True),