Fix RBD level coloring and filter

docs/src/reference/change_log.md

@@ -5,7 +5,10 @@ We also use this change log to document new features that maintain backward comp
 
 ## New features since last version update
 
+- 30 January 2024: Fix RBD-level coloring by updating clade label and clade parsing. [PR 1094](https://github.com/nextstrain/ncov/pull/1094)
+
 - 14 Dec 2023: Use `nextclade2` binary that makes the version explicit [PR 1089](https://github.com/nextstrain/ncov/pull/1089)
+
 - 17 June 2023: Update subsampling strategy for `nextstrain_profiles` to better equilibrate per-capita sampling rates across geographic regions. Primarily this update breaks out China and India as separate subsampling targets because of their large population sizes. It also fine tunes the per-region sampling targets. After this update, URL structure (ie https://nextstrain.org/ncov/gisaid/global/6m) is unchanged. [PR 1074](https://github.com/nextstrain/ncov/pull/1074)
 
 ## v13 (16 May 2023)

scripts/assign_rbd_levels.py

@@ -7,12 +7,12 @@
 def find_matching_nodes(clades_fname, basal_clade_label, tree_fname):
     basal_node_name = None
     with open(clades_fname) as fh:
-        for name, node_data in json.load(fh)['nodes'].items():
-            if node_data.get('clade_annotation', '') == basal_clade_label:
+        for name, node_data in json.load(fh)['branches'].items():
+            if node_data.get('labels', {}).get('clade', '') == basal_clade_label:
                 basal_node_name = name
                 break
     if not basal_node_name:
-        print(f"WARNING: no node found with a clade_annotation of {basal_clade_label}. This script will proceed, but no levels will be exported.")
+        print(f"WARNING: no branch found with a clade of {basal_clade_label}. This script will proceed, but no levels will be exported.")
         return set()
     print(f"Node representing {basal_clade_label}: {basal_node_name}")
     T = Phylo.read(tree_fname, 'newick')
@@ -92,4 +92,4 @@ def classify_into_levels(spike_seq, rbd_mutations):
         node_data['rbd_level_details'][name] = ", ".join([f"S:{x[0][1]}{x[1]} ({x[2]})" for x in zip(rbd_mutations, codons, calls)])
 
     with open(args.output_node_data, 'w') as fh:
-        json.dump(node_data, fh, indent=2)
+        json.dump(node_data, fh, indent=2)

workflow/snakemake_rules/export_for_nextstrain.smk

@@ -303,7 +303,6 @@ rule auspice_config:
                 "pango_lineage",
                 "Nextclade_pango",
                 "region",
-                "level",
                 "country",
                 "division",
                 location_filter,

workflow/snakemake_rules/main_workflow.smk

@@ -1304,11 +1304,13 @@ rule assign_rbd_levels:
         tree = "results/{build_name}/tree.nwk",
     params:
         config=config["files"]["rbd_level_definitions"],
-        basal_clade_label="21L (Omicron)"
+        basal_clade_label="21L (BA.2)"
     output:
         node_data="results/{build_name}/rbd_levels.json",
+    log:
+        "logs/assign_rbd_levels_{build_name}.txt"
     benchmark:
-        "benchmarks/assign_levels_{build_name}.txt",
+        "benchmarks/assign_rbd_levels_{build_name}.txt",
     conda:
         config["conda_environment"],
     shell: