example-open: customize Auspice config and description

Author: joverlee521

profiles/example-open/builds.yaml

@@ -1,3 +1,6 @@
+custom_rules:
+  - profiles/example-open/overwrite_auspice_config.smk
+
 strain_id_field: accession
 
 inputs:
@@ -32,7 +35,9 @@ builds:
       clades: "config/h3n2/ha/clades.tsv"
       subclades: "config/h3n2/{segment}/subclades.tsv"
       lineage: "h3n2"
-      auspice_config: "config/h3n2/{segment}/auspice_config.json"
+      # Created in the workflow with custom rule `overwrite_auspice_config`
+      auspice_config: "data/h3n2/{segment}/auspice_config.json"
+      description: "profiles/example-open/description.md"
       subsamples:
         global:
             filters: "--exclude-where 'ha!=True' 'na!=True' --exclude-ambiguous-dates-by any --subsample-max-sequences 300 --group-by region month --min-date 3Y"

profiles/example-open/description.md

@@ -0,0 +1,8 @@
+We gratefully acknowledge the authors, originating and submitting laboratories
+of the genetic sequences and metadata for sharing their work. This work is made
+possible by the open sharing of genetic data by research groups from all over
+the world. We gratefully acknowledge their contributions.
+
+Special thanks to [GenSpectrum](https://loculus.genspectrum.org) for the
+curation work to link multiple INSDC segment sequences to a single sample and
+for making the data accessible via an open API.

profiles/example-open/overwrite_auspice_config.smk

@@ -0,0 +1,11 @@
+rule overwrite_auspice_config:
+  input:
+    auspice_config = "config/{lineage}/{segment}/auspice_config.json",
+  output:
+    auspice_config = "data/{lineage}/{segment}/auspice_config.json",
+  shell:
+    """
+    jq '.data_provenance = [{{"name": "GenSpectrum", "url": "https://loculus.genspectrum.org"}}]' \
+      {input.auspice_config:q} \
+      > {output.auspice_config:q}
+    """