Merge pull request #254 from nextstrain/james/clade-i

Add Clade-I build

.github/workflows/rebuild-clade-i.yaml

@@ -0,0 +1,59 @@
+name: Rebuild clade-I
+
+on:
+  repository_dispatch:
+    types:
+      - rebuild
+      - rebuild_clade-i
+
+  workflow_dispatch:
+    inputs:
+      trial_name:
+        description: "If set, result will be at nextstrain.org/staging/trial/${trial_name}/${auspice_name}"
+        required: false
+
+jobs:
+  set_config_overrides:
+    runs-on: ubuntu-latest
+    steps:
+      - id: config
+        name: Set config overrides
+        env:
+          TRIAL_NAME: ${{ inputs.trial_name }}
+        run: |
+          config=""
+          if [[ "$TRIAL_NAME" ]]; then
+            config+="--config"
+            config+=" deploy_url='s3://nextstrain-staging/'"
+            config+=" auspice_prefix='"$TRIAL_NAME"'"
+          fi
+
+          echo "config=$config" >> "$GITHUB_OUTPUT"
+    outputs:
+      config_overrides: ${{ steps.config.outputs.config }}
+
+  rebuild_clade_i:
+    needs: [set_config_overrides]
+    permissions:
+      id-token: write
+    uses: nextstrain/.github/.github/workflows/pathogen-repo-build.yaml@master
+    secrets: inherit
+    with:
+      # We can migrate to AWS Batch when/if we need to for more resources,
+      # but at the time of writing the clade-I build is small & quick
+      runtime: docker
+      env: |
+        CONFIG_OVERRIDES: ${{ needs.set_config_overrides.outputs.config_overrides }}
+        GITHUB_RUN_ID: ${{ github.run_id }}
+        SLACK_CHANNELS: ${{ inputs.trial_name && vars.TEST_SLACK_CHANNEL || vars.SLACK_CHANNELS }}
+        BUILD_DIR: phylogenetic
+        BUILD_NAME: clade-i
+      run: |
+        nextstrain build \
+          --env GITHUB_RUN_ID \
+          --env SLACK_TOKEN \
+          --env SLACK_CHANNELS \
+          . \
+            notify_on_deploy \
+            --configfiles $BUILD_DIR/defaults/$BUILD_NAME/config.yaml $BUILD_DIR/build-configs/nextstrain-automation/config.yaml \
+            $CONFIG_OVERRIDES --directory $BUILD_DIR --snakefile $BUILD_DIR/Snakefile

phylogenetic/README.md

@@ -80,6 +80,12 @@ Run pipeline to produce the "lineage B.1" tree for `/mpox/lineage-B.1` with:
 nextstrain build . --configfile defaults/hmpxv1_big/config.yaml
 ```
 
+Run pipeline to produce the "clade I" tree for `/mpox/clade-I` with:
+
+```bash
+nextstrain build . --configfile defaults/clade-i/config.yaml
+```
+
 ### Visualize results
 
 View results with:

phylogenetic/defaults/clade-i/auspice_config.json

@@ -0,0 +1,70 @@
+{
+  "title": "Genomic epidemiology of mpox clade I viruses",
+  "maintainers": [
+    {"name": "Nextstrain team", "url": "http://nextstrain.org"}
+  ],
+  "data_provenance": [
+    {
+      "name": "GenBank",
+      "url": "https://www.ncbi.nlm.nih.gov/genbank/"
+    }
+  ],
+  "build_url": "https://github.com/nextstrain/mpox",
+  "colorings": [
+    {
+      "key": "region",
+      "title": "Region",
+      "type": "categorical"
+    },
+    {
+      "key": "country",
+      "title": "Country",
+      "type": "categorical"
+    },
+    {
+      "key": "host",
+      "title": "Host",
+      "type": "categorical"
+    },
+    {
+      "key": "GA_CT_fraction",
+      "title": "G→A or C→T fraction",
+      "type": "continuous"
+    },
+    {
+      "key": "dinuc_context_fraction",
+      "title": "NGA/TCN context of G→A/C→T mutations",
+      "type": "continuous"
+    },
+    {
+      "key": "recency",
+      "title": "Submission Recency",
+      "type": "categorical"
+    },
+    {
+      "key": "date_submitted",
+      "title": "Release Date",
+      "type": "categorical"
+    },
+    {
+      "key": "date",
+      "title": "Collection date",
+      "type": "categorical"
+    }
+  ],
+  "geo_resolutions": [
+    "country"
+  ],
+  "display_defaults": {
+    "color_by": "country",
+    "map_triplicate": true,
+    "distance_measure": "num_date",
+    "transmission_lines": false
+  },
+  "filters": [
+    "country",
+    "region",
+    "recency",
+    "host"
+  ]
+}

phylogenetic/defaults/clade-i/config.yaml

@@ -0,0 +1,58 @@
+reference: "defaults/reference.fasta"
+genome_annotation: "defaults/genome_annotation.gff3"
+genbank_reference: "defaults/reference.gb"
+include: "defaults/clade-i/include.txt"
+clades: "defaults/clades.tsv"
+lat_longs: "defaults/lat_longs.tsv"
+auspice_config: "defaults/clade-i/auspice_config.json"
+description: "defaults/description.md"
+tree_mask: "defaults/tree_mask.tsv"
+
+# Use `accession` as the ID column since `strain` currently contains duplicates¹.
+# ¹ https://github.com/nextstrain/mpox/issues/33
+strain_id_field: "accession"
+display_strain_field: "strain"
+
+build_name: "clade-i"
+auspice_name: "mpox_clade-I"
+
+filter:
+  min_date: 1900
+  min_length: 100000
+  exclude_where: 'clade!=I'
+
+
+### We don't want to subsample, so specify a config which is essentially a no-op
+subsample:
+  everything:
+    group_by: ""
+    sequences_per_group: ""
+
+## align
+max_indel: 10000
+seed_spacing: 1000
+
+## treefix
+fix_tree: true
+treefix_root: "" # without a root we'll midpoint root which should work great for clade I
+
+## refine
+timetree: true
+root: "best"
+# Clock rate chosen via treetime inference on Clade-I data excluding Clade-Ib seqs (n=73)
+# TODO: update this once more public data is available.
+clock_rate: 1.465e-06
+clock_std_dev: 6.7e-07
+divergence_units: "mutations"
+
+traits:
+  columns: "country"
+  sampling_bias_correction: 3
+
+## recency
+recency: true
+
+mask:
+  from_beginning: 800
+  from_end: 6422
+  maskfile: "defaults/mask.bed"

phylogenetic/defaults/clades.tsv

@@ -5,6 +5,9 @@ outgroup	nuc	179226	T
 clade I	nuc	87560	T
 clade I	nuc	136015	A
 
+clade Ib	nuc	6014	G
+clade Ib	nuc	108966	T
+
 clade II	nuc	86502	G
 clade II	nuc	150970	A
 clade II	nuc	35352	C

phylogenetic/rules/prepare_sequences.smk

@@ -65,6 +65,11 @@ rule filter:
         min_date=config["filter"]["min_date"],
         min_length=config["filter"]["min_length"],
         strain_id=config["strain_id_field"],
+        exclude_where=lambda w: (
+            f"--exclude-where {config['filter']['exclude_where']}"
+            if "exclude_where" in config["filter"]
+            else ""
+        ),
     shell:
         """
         augur filter \
@@ -74,6 +79,7 @@ rule filter:
             --output-sequences {output.sequences} \
             --output-metadata {output.metadata} \
             --exclude {input.exclude} \
+            {params.exclude_where} \
             --min-date {params.min_date} \
             --min-length {params.min_length} \
             --query "(QC_rare_mutations == 'good' | QC_rare_mutations == 'mediocre')" \