Use augur subsample

.gitattributes

@@ -1,3 +1,8 @@
 # Allow Git to decide if file is text or binary
 # Always use LF line endings even on Windows.
 * text=auto eol=lf
+
+# This is a large generated file that, while text, it is not useful to
+# routinely show the diff of.  A diff can be forced as needed, e.g. with `git
+# diff --text`.
+/config/configfile.yaml -diff

CHANGELOG.md

@@ -1,3 +1,9 @@
 # CHANGELOG
 
 We use this CHANGELOG to document breaking changes, new features, bug fixes, and config value changes that may affect both the usage of the workflows and the outputs of the workflows.
+
+## 2026
+
+* TBD: The `filter` section in phylogenetic workflow configuration has been replaced by `subsample`/`custom_subsample` for subsampling, and `filter_for_f_antibody_escape` for initial quality filtering. **This is a breaking change**.
+    * NOTE: The workflow does not yet support proximal samples.
+* TBD: Phylogenetic workflow configuration is now validated against a strict schema. The workflow will error if your configuration has extraneous entries that were previously ignored.

README.md

@@ -46,7 +46,14 @@ Once you've run the build, you can view the results with:
 
 ## Configuration
 
-The default configuration is in [`config/configfile.yaml`](./config/configfile.yaml).
+The default configuration is generated by
+[scripts/generate_default_config.py](./scripts/generate_default_config.py). It
+can be run with Nextstrain CLI:
+
+```sh
+nextstrain shell . -c 'python scripts/generate_default_config.py'
+```
+
 The workflow is contained in the [Snakefile](Snakefile) with included
 [rules](workflow/snakemake_rules/). Each rule specifies its file inputs and outputs
 and pulls its parameters from the config. There is little redirection and each

config.schema.yaml

@@ -0,0 +1,227 @@
+$schema: "https://json-schema.org/draft/2020-12/schema"
+title: RSV Phylogenetic Workflow Configuration
+description: >-
+  This is the schema for the Nextstrain rsv phylogenetic workflow's
+  configuration file.
+
+$defs:
+  per_full_build_map: &per_full_build_map
+    type: object
+    additionalProperties: false
+    propertyNames:
+      title: Full build name
+      description: Full build name (e.g. 'a/genome/all-time')
+
+  per_subtype_map: &per_subtype_map
+    type: object
+    additionalProperties: false
+    propertyNames:
+      title: Subtype
+      description: Subtype ('a' or 'b')
+
+  per_build_map: &per_build_map
+    type: object
+    additionalProperties: false
+    propertyNames:
+      title: Build name
+      description: Build name ('genome', 'G', 'F', or 'F-antibody-escape')
+
+  per_resolution_map: &per_resolution_map
+    type: object
+    additionalProperties: false
+    propertyNames:
+      title: Resolution name
+      description: Resolution name ('all-time', '6y', or '3y')
+
+  input_item:
+    type: object
+    additionalProperties: false
+    required:
+      - name
+    anyOf:
+      - required: [metadata]
+      - required: [sequences]
+    properties:
+      name:
+        type: string
+      metadata:
+        type: string
+      sequences:
+        type: string
+
+type: object
+additionalProperties: false
+properties:
+  conda_environment:
+    type: string
+  genesforglycosylation:
+    type: array
+    items:
+      type: string
+  builds_to_run:
+    type: array
+    items:
+      type: string
+  resolutions_to_run:
+    type: array
+    items:
+      type: string
+  subtypes:
+    type: array
+    items:
+      type: string
+  inputs:
+    type: array
+    items:
+      $ref: "#/$defs/input_item"
+  additional_inputs:
+    type: array
+    items:
+      $ref: "#/$defs/input_item"
+  exclude:
+    type: string
+  description:
+    type: string
+  strain_id_field:
+    type: string
+  display_strain_field:
+    type: string
+  subsample: &subsample_config
+    <<: *per_full_build_map
+    description: >-
+      Subsampling configuration. When using --configfile, it is recommended to
+      use 'custom_subsample' instead to ignore default subsampling configuration.
+    patternProperties:
+      "^.*$":
+        $ref: "https://nextstrain.org/schemas/augur/subsample-config/v1#/$defs/schemaForUnalignedSequences"
+  custom_subsample:
+    <<: *subsample_config
+    description: >-
+      Custom subsampling configuration. When using --configfile, this is
+      recommended over 'subsample' to ignore default subsampling configuration.
+  files:
+    type: object
+    additionalProperties: false
+    properties:
+      auspice_config:
+        type: string
+      auspice_config_additional_colorings:
+        type: string
+      auspice_config_f_antibody_escape:
+        type: string
+      auspice_config_non-genome_builds:
+        type: string
+  refine:
+    type: object
+    additionalProperties: false
+    properties:
+      coalescent:
+        type: string
+      date_inference:
+        type: string
+      clock_filter_iqd:
+        type: number
+      divergence_units:
+        type: string
+  ancestral:
+    type: object
+    additionalProperties: false
+    properties:
+      inference:
+        type: string
+  cds:
+    <<: *per_build_map
+    patternProperties:
+      "^.*$":
+        type: string
+  traits:
+    type: object
+    additionalProperties: false
+    properties:
+      columns:
+        type: [string, array]
+        items:
+          type: string
+  frequencies:
+    type: object
+    additionalProperties: false
+    properties:
+      resolutions:
+        <<: *per_resolution_map
+        patternProperties:
+          "^.*$":
+            type: object
+            additionalProperties: false
+            properties:
+              min_date:
+                type: string
+  nextclade_attributes:
+    <<: *per_subtype_map
+    patternProperties:
+      "^.*$":
+        type: object
+        additionalProperties: false
+        properties:
+          name:
+            type: string
+          reference_name:
+            type: string
+          accession:
+            type: string
+  filter_for_f_antibody_escape:
+    type: object
+    additionalProperties: false
+    properties:
+      group_by:
+        type: string
+      min_length:
+        <<: *per_build_map
+        patternProperties:
+          "^.*$":
+            type: integer
+      min_coverage:
+        <<: *per_build_map
+        patternProperties:
+          "^.*$":
+            type: number
+      resolutions:
+        <<: *per_resolution_map
+        patternProperties:
+          "^.*$":
+            type: object
+            additionalProperties: false
+            properties:
+              min_date:
+                type: string
+              background_min_date:
+                type: string
+  f_dms_data:
+    type: string
+  f_dms_antibodies:
+    type: array
+    items:
+      type: string
+  dms_only_positive_escape:
+    type: boolean
+  enrich_antibody_escape:
+    <<: *per_build_map
+    patternProperties:
+      "^.*$":
+        type: object
+        additionalProperties: false
+        properties:
+          nseqs_per_antibody_scoretype:
+            type: integer
+          group_by:
+            type: array
+            items:
+              type: string
+          max_identical_f_prot_muts:
+            type: integer
+          max_identical_max_escape_mut:
+            type: integer
+  custom_rules:
+    type: array
+    description: Custom Snakemake rule files to include. If used, this will disable config schema validation.
+    items:
+      type: string