Phylogenetic: Use group-by with subsample-max-sequences

nextstrain · Jan 14, 2025 · 0d73b74 · 0d73b74
1 parent 05b014e
commit 0d73b74
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 8 deletions.
diff --git a/phylogenetic/defaults/config_dengue.yaml b/phylogenetic/defaults/config_dengue.yaml
@@ -14,12 +14,7 @@ filter:
   min_length:
     genome: 5000
     E: 1000
-  sequences_per_group:
-    all: '36'
-    denv1: '36'
-    denv2: '36'
-    denv3: '36'
-    denv4: '36'
+  subsample_max_sequences: '4000'
 
 traits:
   sampling_bias_correction: '3'

diff --git a/phylogenetic/rules/prepare_sequences.smk b/phylogenetic/rules/prepare_sequences.smk
@@ -60,7 +60,7 @@ rule filter:
         sequences = "results/{gene}/filtered_{serotype}.fasta"
     params:
         group_by = config['filter']['group_by'],
-        sequences_per_group = lambda wildcards: config['filter']['sequences_per_group'][wildcards.serotype],
+        subsample_max_sequences = config['filter']['subsample_max_sequences'],
         min_length = lambda wildcard: config['filter']['min_length'][wildcard.gene],
         strain_id = config.get("strain_id_field", "strain"),
     shell:
@@ -73,7 +73,7 @@ rule filter:
             --include {input.include} \
             --output {output.sequences} \
             --group-by {params.group_by} \
-            --sequences-per-group {params.sequences_per_group} \
+            --subsample-max-sequences {params.subsample_max_sequences} \
             --min-length {params.min_length} \
             --exclude-where country=? region=? date=? is_lab_host='true' \
             --query-columns is_lab_host:str