phylogenetic: Use augur subsample

CHANGELOG.md

@@ -10,6 +10,7 @@ Instead, changes appear below grouped by the date they were added to the workflo
 
 ## 2025
 
+* TBD: [breaking] Switch to use `augur subsample`.
 * 02 July 2025: phylogenetic - config schema updates for easier config overlays ([#321][])
     * new required config params
         * `exclude` - path to exclude.txt for `augur filter`

phylogenetic/Snakefile

@@ -2,7 +2,7 @@ from packaging import version
 from augur.__version__ import __version__ as augur_version
 import sys
 
-min_augur_version = "22.2.0"
+min_augur_version = "31.5.0"
 if version.parse(augur_version) < version.parse(min_augur_version):
     print("This pipeline needs a newer version of augur than you currently have...")
     print(
@@ -39,6 +39,7 @@ rule all:
 
 
 include: "rules/config.smk"
+include: "rules/write_config.smk"
 include: "rules/prepare_sequences.smk"
 include: "rules/construct_phylogeny.smk"
 include: "rules/annotate_phylogeny.smk"

phylogenetic/build-configs/ci/config.yaml

@@ -4,7 +4,6 @@ custom_rules:
 reference: "defaults/reference.fasta"
 genome_annotation: "defaults/genome_annotation.gff3"
 genbank_reference: "defaults/reference.gb"
-include: "defaults/hmpxv1/include.txt"
 exclude: "defaults/exclude.txt"
 clades: "defaults/clades.tsv"
 lat_longs: "defaults/lat_longs.tsv"
@@ -30,52 +29,62 @@ filter:
 ### Set 1: Non-B.1 sequences: use all
 ### Set 2: B.1 sequences: small sample across year/country, maybe month
 subsample:
-  non_b1: >-
-    --group-by lineage year country
-    --sequences-per-group 50
-    --exclude-where
-    outbreak!=hMPXV-1
-    clade!=IIb
-    lineage=B.1
-    lineage=B.1.1
-    lineage=B.1.2
-    lineage=B.1.3
-    lineage=C.1
-    lineage=C.1.1
-    lineage=E.1
-    lineage=E.2
-    lineage=E.3
-    lineage=B.1.4
-    lineage=B.1.5
-    lineage=B.1.6
-    lineage=D.1
-    lineage=B.1.7
-    lineage=B.1.8
-    lineage=B.1.9
-    lineage=B.1.10
-    lineage=B.1.11
-    lineage=B.1.12
-    lineage=B.1.13
-    lineage=B.1.14
-    lineage=B.1.15
-    lineage=B.1.16
-    lineage=B.1.17
-    lineage=B.1.18
-    lineage=B.1.19
-    lineage=B.1.20
-    lineage=F.1
-    lineage=F.2
-    lineage=F.3
-    lineage=F.4
-    lineage=F.5
-    lineage=F.6
-    lineage=B.1.21
-    lineage=B.1.22
-    lineage=B.1.23
-  b1: >-
-    --group-by country year
-    --subsample-max-sequences 300
-    --exclude-where outbreak!=hMPXV-1 clade!=IIb
+  samples:
+    non_b1:
+      group_by:
+        - lineage
+        - year
+        - country
+      sequences_per_group: 50
+      exclude_where:
+        - outbreak!=hMPXV-1
+        - clade!=IIb
+        - lineage=B.1
+        - lineage=B.1.1
+        - lineage=B.1.2
+        - lineage=B.1.3
+        - lineage=C.1
+        - lineage=C.1.1
+        - lineage=E.1
+        - lineage=E.2
+        - lineage=E.3
+        - lineage=B.1.4
+        - lineage=B.1.5
+        - lineage=B.1.6
+        - lineage=D.1
+        - lineage=B.1.7
+        - lineage=B.1.8
+        - lineage=B.1.9
+        - lineage=B.1.10
+        - lineage=B.1.11
+        - lineage=B.1.12
+        - lineage=B.1.13
+        - lineage=B.1.14
+        - lineage=B.1.15
+        - lineage=B.1.16
+        - lineage=B.1.17
+        - lineage=B.1.18
+        - lineage=B.1.19
+        - lineage=B.1.20
+        - lineage=F.1
+        - lineage=F.2
+        - lineage=F.3
+        - lineage=F.4
+        - lineage=F.5
+        - lineage=F.6
+        - lineage=B.1.21
+        - lineage=B.1.22
+        - lineage=B.1.23
+      include:
+        - defaults/hmpxv1/include.txt
+    b1:
+      group_by:
+        - country
+        - year
+      max_sequences: 300
+      exclude_where:
+        - outbreak!=hMPXV-1
+        - clade!=IIb
 
 ## align
 max_indel: 10000

phylogenetic/build-configs/inrb/config.yaml

@@ -21,5 +21,5 @@ traits:
 # Private INRB data doesn't have clade annotations so allow empty clade fields
 # (i.e. we're assuming all INRB data is clade I)
 subsample:
-  everything: >-
-    --query 'clade in ["I", "Ia", "Ib", ""]'
+  everything:
+    query: clade in ["I", "Ia", "Ib", ""]

phylogenetic/defaults/clade-i/config.yaml

@@ -1,7 +1,6 @@
 reference: "defaults/clade-i/reference.fasta"
 genome_annotation: "defaults/clade-i/genome_annotation.gff3"
 genbank_reference: "defaults/clade-i/reference.gb"
-include: "defaults/clade-i/include.txt"
 exclude: "defaults/exclude.txt"
 clades: "defaults/clades.tsv"
 lat_longs: "defaults/lat_longs.tsv"
@@ -26,8 +25,11 @@ filter:
 
 ### Filter to only Clade I sequences
 subsample:
-  everything: >-
-    --query 'clade in ["I", "Ia", "Ib"]'
+  samples:
+    everything:
+      query: clade in ["I", "Ia", "Ib"]
+      include:
+        - defaults/clade-i/include.txt
 
 ## align
 max_indel: 10000

phylogenetic/defaults/hmpxv1/config.yaml

@@ -1,7 +1,6 @@
 reference: "defaults/reference.fasta"
 genome_annotation: "defaults/genome_annotation.gff3"
 genbank_reference: "defaults/reference.gb"
-include: "defaults/hmpxv1/include.txt"
 exclude: "defaults/exclude.txt"
 clades: "defaults/clades.tsv"
 lat_longs: "defaults/lat_longs.tsv"
@@ -27,52 +26,62 @@ filter:
 ### Set 1: Non-B.1 sequences: use all
 ### Set 2: B.1 sequences: small sample across year/country, maybe month
 subsample:
-  non_b1: >-
-    --group-by lineage year country
-    --sequences-per-group 50
-    --exclude-where
-    outbreak!=hMPXV-1
-    clade!=IIb
-    lineage=B.1
-    lineage=B.1.1
-    lineage=B.1.2
-    lineage=B.1.3
-    lineage=C.1
-    lineage=C.1.1
-    lineage=E.1
-    lineage=E.2
-    lineage=E.3
-    lineage=B.1.4
-    lineage=B.1.5
-    lineage=B.1.6
-    lineage=D.1
-    lineage=B.1.7
-    lineage=B.1.8
-    lineage=B.1.9
-    lineage=B.1.10
-    lineage=B.1.11
-    lineage=B.1.12
-    lineage=B.1.13
-    lineage=B.1.14
-    lineage=B.1.15
-    lineage=B.1.16
-    lineage=B.1.17
-    lineage=B.1.18
-    lineage=B.1.19
-    lineage=B.1.20
-    lineage=F.1
-    lineage=F.2
-    lineage=F.3
-    lineage=F.4
-    lineage=F.5
-    lineage=F.6
-    lineage=B.1.21
-    lineage=B.1.22
-    lineage=B.1.23
-  b1: >-
-    --group-by country year
-    --subsample-max-sequences 300
-    --exclude-where outbreak!=hMPXV-1 clade!=IIb
+  samples:
+    non_b1:
+      group_by:
+        - lineage
+        - year
+        - country
+      sequences_per_group: 50
+      exclude_where:
+        - outbreak!=hMPXV-1
+        - clade!=IIb
+        - lineage=B.1
+        - lineage=B.1.1
+        - lineage=B.1.2
+        - lineage=B.1.3
+        - lineage=C.1
+        - lineage=C.1.1
+        - lineage=E.1
+        - lineage=E.2
+        - lineage=E.3
+        - lineage=B.1.4
+        - lineage=B.1.5
+        - lineage=B.1.6
+        - lineage=D.1
+        - lineage=B.1.7
+        - lineage=B.1.8
+        - lineage=B.1.9
+        - lineage=B.1.10
+        - lineage=B.1.11
+        - lineage=B.1.12
+        - lineage=B.1.13
+        - lineage=B.1.14
+        - lineage=B.1.15
+        - lineage=B.1.16
+        - lineage=B.1.17
+        - lineage=B.1.18
+        - lineage=B.1.19
+        - lineage=B.1.20
+        - lineage=F.1
+        - lineage=F.2
+        - lineage=F.3
+        - lineage=F.4
+        - lineage=F.5
+        - lineage=F.6
+        - lineage=B.1.21
+        - lineage=B.1.22
+        - lineage=B.1.23
+      include:
+        - defaults/hmpxv1/include.txt
+    b1:
+      group_by:
+        - country
+        - year
+      max_sequences: 300
+      exclude_where:
+        - outbreak!=hMPXV-1
+        - clade!=IIb
 
 ## align
 max_indel: 10000

phylogenetic/defaults/hmpxv1_big/config.yaml

@@ -1,7 +1,6 @@
 reference: "defaults/reference.fasta"
 genome_annotation: "defaults/genome_annotation.gff3"
 genbank_reference: "defaults/reference.gb"
-include: "defaults/hmpxv1_big/include.txt"
 exclude: "defaults/exclude.txt"
 clades: "defaults/clades.tsv"
 lat_longs: "defaults/lat_longs.tsv"
@@ -24,20 +23,26 @@ filter:
   query: "(QC_rare_mutations == 'good' | QC_rare_mutations == 'mediocre')"
 
 subsample:
-  b1: >-
-    --group-by year month country
-    --subsample-max-sequences 5000
-    --exclude-where
-    outbreak!=hMPXV-1
-    clade!=IIb
-    lineage=A
-    lineage=A.1
-    lineage=A.1.1
-    lineage=A.2
-    lineage=A.2.1
-    lineage=A.2.2
-    lineage=A.2.3
-    lineage=A.3
+  samples:
+    b1:
+      group_by:
+        - year
+        - month
+        - country
+      max_sequences: 5000
+      exclude_where:
+        - outbreak!=hMPXV-1
+        - clade!=IIb
+        - lineage=A
+        - lineage=A.1
+        - lineage=A.1.1
+        - lineage=A.2
+        - lineage=A.2.1
+        - lineage=A.2.2
+        - lineage=A.2.3
+        - lineage=A.3
+      include:
+        - defaults/hmpxv1_big/include.txt
 
 ## align
 max_indel: 10000