[build] Update zika build to pull new ingest data

Since Ingest pushes data to a different endpoint, update zika build to pull from the new endpoint. Other modifications from the original described below: Since strains (or isolates) may be re-sequenced resulting in duplicate strain names breaking Nextstrain builds, other pathogen repos (e.g. Monkeypox) had switched to indexing records by GenBank ID instead (`rule wrangle_metdata`) and swapping in the final strain name at the end (`rule final_strain_name`) and Zika was updated accordingly. Other changes include updating the list of dropped strain names to GenBank IDs (where the original strain names moved to comments) and updating the example sequence fasta headers. In addition, added the strain name as a Tip Label dropdown item as discussed in: #25 (comment)
nextstrain · May 4, 2023 · 9f4f694 · 9f4f694
1 parent 31f044f
commit 9f4f694
Show file tree

Hide file tree

Showing 6 changed files with 211 additions and 128 deletions.
diff --git a/Snakefile b/Snakefile
@@ -1,3 +1,6 @@
+if not config:
+    configfile: "config/config_zika.yaml"
+
 rule all:
     input:
         auspice_json = "auspice/zika.json",
@@ -17,10 +20,10 @@ rule download:
     message: "Downloading sequences and metadata from data.nextstrain.org"
     output:
         sequences = "data/sequences.fasta.zst",
-        metadata = "data/metadata.tsv.zst"
+        metadata = "data/metadata.tsv.zst",
     params:
-        sequences_url = "https://data.nextstrain.org/files/zika/sequences.fasta.zst",
-        metadata_url = "https://data.nextstrain.org/files/zika/metadata.tsv.zst"
+        sequences_url = "https://data.nextstrain.org/files/workflows/zika/sequences_all.fasta.zst",
+        metadata_url = "https://data.nextstrain.org/files/workflows/zika/metadata_all.tsv.zst"
     shell:
         """
         curl -fsSL --compressed {params.sequences_url:q} --output {output.sequences}
@@ -30,8 +33,8 @@ rule download:
 rule decompress:
     message: "Decompressing sequences and metadata"
     input:
-        sequences = "data/sequences.fasta.zst",
-        metadata = "data/metadata.tsv.zst"
+        sequences = "data/sequences_all.fasta.zst",
+        metadata = "data/metadata_all.tsv.zst"
     output:
         sequences = "data/sequences.fasta",
         metadata = "data/metadata.tsv"
@@ -41,6 +44,20 @@ rule decompress:
         zstd -d -c {input.metadata} > {output.metadata}
         """
 
+rule wrangle_metadata:
+    input:
+        metadata="data/metadata.tsv"
+    output:
+        metadata="results/wrangled_metadata.tsv",
+    params:
+        strain_id=lambda w: config.get("strain_id_field", "strain"),
+    shell:
+        """
+        csvtk -t rename -f strain -n strain_original {input.metadata} \
+          | csvtk -t mutate -f {params.strain_id} -n strain > {output.metadata}
+        """
+
+
 rule filter:
     message:
         """
@@ -51,8 +68,8 @@ rule filter:
           - minimum genome length of {params.min_length} (50% of Zika virus genome)
         """
     input:
-        sequences = rules.decompress.output.sequences,
-        metadata = rules.decompress.output.metadata,
+        sequences = "data/sequences.fasta",
+        metadata = rules.wrangle_metadata.output.metadata,
         exclude = files.dropped_strains
     output:
         sequences = "results/filtered.fasta"
@@ -120,7 +137,7 @@ rule refine:
     input:
         tree = rules.tree.output.tree,
         alignment = rules.align.output,
-        metadata = rules.decompress.output.metadata
+        metadata = rules.wrangle_metadata.output.metadata
     output:
         tree = "results/tree.nwk",
         node_data = "results/branch_lengths.json"
@@ -186,7 +203,7 @@ rule traits:
         """
     input:
         tree = rules.refine.output.tree,
-        metadata = rules.decompress.output.metadata
+        metadata = rules.wrangle_metadata.output.metadata
     output:
         node_data = "results/traits.json",
     params:
@@ -207,7 +224,7 @@ rule export:
     message: "Exporting data files for for auspice"
     input:
         tree = rules.refine.output.tree,
-        metadata = rules.decompress.output.metadata,
+        metadata = rules.wrangle_metadata.output.metadata,
         branch_lengths = rules.refine.output.node_data,
         traits = rules.traits.output.node_data,
         nt_muts = rules.ancestral.output.node_data,
@@ -216,7 +233,8 @@ rule export:
         auspice_config = files.auspice_config,
         description = files.description
     output:
-        auspice_json = rules.all.input.auspice_json
+        auspice_json = "results/raw_zika.json",
+        root_sequence = "results/raw_zika_root-sequence.json",
     shell:
         """
         augur export v2 \
@@ -230,6 +248,27 @@ rule export:
             --output {output.auspice_json}
         """
 
+rule final_strain_name:
+    input:
+        auspice_json=rules.export.output.auspice_json,
+        metadata=rules.wrangle_metadata.output.metadata,
+        root_sequence=rules.export.output.root_sequence,
+    output:
+        auspice_json=rules.all.input.auspice_json,
+        root_sequence="auspice/zika_root-sequence.json",
+    params:
+        display_strain_field=lambda w: config.get("display_strain_field", "strain"),
+    shell:
+        """
+        python3 bin/set_final_strain_name.py \
+            --metadata {input.metadata} \
+            --input-auspice-json {input.auspice_json} \
+            --display-strain-name {params.display_strain_field} \
+            --output {output.auspice_json}
+
+        cp {input.root_sequence} {output.root_sequence}
+        """
+
 rule clean:
     message: "Removing directories: {params}"
     params:

diff --git a/bin/set_final_strain_name.py b/bin/set_final_strain_name.py
@@ -0,0 +1,36 @@
+import pandas as pd
+import json, argparse
+
+def replace_name_recursive(node, lookup):
+    if node["name"] in lookup:
+        node["name"] = lookup[node["name"]]
+
+    if "children" in node:
+        for child in node["children"]:
+            replace_name_recursive(child, lookup)
+
+if __name__=="__main__":
+    parser = argparse.ArgumentParser(
+        description="Swaps out the strain names in the Auspice JSON with the final strain name",
+        formatter_class=argparse.ArgumentDefaultsHelpFormatter
+    )
+
+    parser.add_argument('--input-auspice-json', type=str, required=True, help="input auspice_json")
+    parser.add_argument('--metadata', type=str, required=True, help="input data")
+    parser.add_argument('--display-strain-name', type=str, required=True, help="field to use as strain name in auspice")
+    parser.add_argument('--output', type=str, metavar="JSON", required=True, help="output Auspice JSON")
+    args = parser.parse_args()
+
+    metadata = pd.read_csv(args.metadata, sep='\t')
+    name_lookup = {}
+    for ri, row in metadata.iterrows():
+        strain_id = row['strain']
+        name_lookup[strain_id] = args.display_strain_name if pd.isna(row[args.display_strain_name]) else row[args.display_strain_name]
+
+    with open(args.input_auspice_json, 'r') as fh:
+        data = json.load(fh)
+
+    replace_name_recursive(data['tree'], name_lookup)
+
+    with open(args.output, 'w') as fh:
+        json.dump(data, fh)
diff --git a/config/auspice_config.json b/config/auspice_config.json
@@ -5,6 +5,11 @@
   ],
   "build_url": "https://github.com/nextstrain/zika",
   "colorings": [
+    {
+      "key": "strain_original",
+      "title": "strain name",
+      "type": "categorical"
+    },
     {
       "key": "gt",
       "title": "genotype",

diff --git a/config/config_zika.yaml b/config/config_zika.yaml
@@ -0,0 +1,2 @@
+strain_id_field: "accession"
+display_strain_field: "strain_original"
diff --git a/config/dropped_strains.txt b/config/dropped_strains.txt
@@ -1,86 +1,87 @@
-PF13/251013_18 # reference included in config/zika_reference.gb
-AFMC_U # too basal
-AFMC_S # too basal
-Boracay/16423 # too basal
-JMB_185 # too basal
-PHL/2012/CPC_0740 # too basal
+MG827392
+KX369547 # PF13/251013_18 # reference included in config/zika_reference.gb
+KY553111 # AFMC_U # too basal
+KY962729 # AFMC_S # too basal
+KY120353 # Boracay/16423 # too basal
+KU179098 # JMB_185 # too basal
+KU681082 # PHL/2012/CPC_0740 # too basal
 VIE/Bra/2016 # too basal
-Dominican_Republic/2016/PD2 # duplicate of other strain in dataset
-GD01 # duplicate of other strain in dataset
-GDZ16001 # duplicate of other strain in dataset
-VEN/UF_2/2016 # duplicate of other strain in dataset
-ZZ_1 # duplicate of other strain in dataset
-VR10599/Pavia/2016 # export with unknown origin
-34997/Pavia/2016 # export with unknown origin
-COL/FLR_00001/2015 # duplicate of COL/FLR/2015
-COL/FLR_00002/2015 # duplicate of COL/FLR/2015
-COL/FLR_00003/2015 # duplicate of COL/FLR/2015
-COL/FLR_00004/2015 # duplicate of COL/FLR/2015
-COL/FLR_00005/2015 # duplicate of COL/FLR/2015
-COL/FLR_00006/2015 # duplicate of COL/FLR/2015
-COL/FLR_00007/2015 # duplicate of COL/FLR/2015
-COL/FLR_00008/2015 # duplicate of COL/FLR/2015
-COL/FLR_00009/2015 # duplicate of COL/FLR/2015
-COL/FLR_00010/2015 # duplicate of COL/FLR/2015
-COL/FLR_00011/2015 # duplicate of COL/FLR/2015
-COL/FLR_00012/2015 # duplicate of COL/FLR/2015
-COL/FLR_00013/2015 # duplicate of COL/FLR/2015
-COL/FLR_00014/2015 # duplicate of COL/FLR/2015
-COL/FLR_00015/2015 # duplicate of COL/FLR/2015
-COL/FLR_00016/2015 # duplicate of COL/FLR/2015
-COL/FLR_00017/2015 # duplicate of COL/FLR/2015
-COL/FLR_00018/2015 # duplicate of COL/FLR/2015
-COL/FLR_00019/2015 # duplicate of COL/FLR/2015
-COL/FLR_00020/2015 # duplicate of COL/FLR/2015
-COL/FLR_00021/2015 # duplicate of COL/FLR/2015
-COL/FLR_00022/2015 # duplicate of COL/FLR/2015
-COL/FLR_00023/2015 # duplicate of COL/FLR/2015
-COL/FLR_00024/2015 # duplicate of COL/FLR/2015
-COL/FLR_00025/2015 # duplicate of COL/FLR/2015
-COL/FLR_00026/2015 # duplicate of COL/FLR/2015
-COL/FLR_00034/2015 # duplicate of COL/FLR/2015
-COL/FLR_00035/2015 # duplicate of COL/FLR/2015
-COL/FLR_00036/2015 # duplicate of COL/FLR/2015
-COL/FLR_00038/2015 # duplicate of COL/FLR/2015
-COL/FLR_00040/2015 # duplicate of COL/FLR/2015
-COL/FLR_00041/2015 # duplicate of COL/FLR/2015
-COL/FLR_00042/2015 # duplicate of COL/FLR/2015
-COL/PRV_00027/2015 # misdated
-COL/PRV_00028/2015 # misdated
-COL/PAN_00029/2015 # misdated
-COL/PAN_00030/2015 # misdated
-BRA/2016/FC_DQ12D1 # large indel
-Brazil/2016/ZBRX8 # large indel
-Brazil/2016/ZBRX11 # large indel
-CX17 # large indel
-MEX/2016/mex27 # large indel
-MEX/2016/mex50 # large indel
-SLV/2016/ElSalvador_1055 # large indel
-USVI/20/2016 # large indel
+KU853013 # Dominican_Republic/2016/PD2 # duplicate of other strain in dataset
+KU740184 # GD01 # duplicate of other strain in dataset
+KU761564 # GDZ16001 # duplicate of other strain in dataset
+KX893855 # VEN/UF_2/2016 # duplicate of other strain in dataset
+KY927808 # ZZ_1 # duplicate of other strain in dataset
+KY003154 # VR10599/Pavia/2016 # export with unknown origin
+KY003153 # 34997/Pavia/2016 # export with unknown origin
+MF574552 # COL/FLR_00001/2015 # duplicate of COL/FLR/2015
+MF574559 # COL/FLR_00002/2015 # duplicate of COL/FLR/2015
+MF574560 # COL/FLR_00003/2015 # duplicate of COL/FLR/2015
+MF574561 # COL/FLR_00004/2015 # duplicate of COL/FLR/2015
+MF574571 # COL/FLR_00005/2015 # duplicate of COL/FLR/2015
+MF574555 # COL/FLR_00006/2015 # duplicate of COL/FLR/2015
+MF574557 # COL/FLR_00007/2015 # duplicate of COL/FLR/2015
+MF574562 # COL/FLR_00008/2015 # duplicate of COL/FLR/2015
+MF574572 # COL/FLR_00009/2015 # duplicate of COL/FLR/2015
+MF574570 # COL/FLR_00010/2015 # duplicate of COL/FLR/2015
+MF574565 # COL/FLR_00011/2015 # duplicate of COL/FLR/2015
+MF574568 # COL/FLR_00012/2015 # duplicate of COL/FLR/2015
+MF574558 # COL/FLR_00013/2015 # duplicate of COL/FLR/2015
+MF574576 # COL/FLR_00014/2015 # duplicate of COL/FLR/2015
+MF574567 # COL/FLR_00015/2015 # duplicate of COL/FLR/2015
+MF574575 # COL/FLR_00016/2015 # duplicate of COL/FLR/2015
+MF574553 # COL/FLR_00017/2015 # duplicate of COL/FLR/2015
+MF574573 # COL/FLR_00018/2015 # duplicate of COL/FLR/2015
+MF574574 # COL/FLR_00019/2015 # duplicate of COL/FLR/2015
+MF574577 # COL/FLR_00020/2015 # duplicate of COL/FLR/2015
+MF574556 # COL/FLR_00021/2015 # duplicate of COL/FLR/2015
+MF574554 # COL/FLR_00022/2015 # duplicate of COL/FLR/2015
+MF574566 # COL/FLR_00023/2015 # duplicate of COL/FLR/2015
+MF574569 # COL/FLR_00024/2015 # duplicate of COL/FLR/2015
+MF574563 # COL/FLR_00025/2015 # duplicate of COL/FLR/2015
+MF574564 # COL/FLR_00026/2015 # duplicate of COL/FLR/2015
+MF574581 # COL/FLR_00034/2015 # duplicate of COL/FLR/2015
+MF574588 # COL/FLR_00035/2015 # duplicate of COL/FLR/2015
+MF574582 # COL/FLR_00036/2015 # duplicate of COL/FLR/2015
+MF574586 # COL/FLR_00038/2015 # duplicate of COL/FLR/2015
+MF574584 # COL/FLR_00040/2015 # duplicate of COL/FLR/2015
+MF574583 # COL/FLR_00041/2015 # duplicate of COL/FLR/2015
+MF574580 # COL/FLR_00042/2015 # duplicate of COL/FLR/2015
+MF574579 # COL/PRV_00027/2015 # misdated
+MF574578 # COL/PRV_00028/2015 # misdated
+MF574585 # COL/PAN_00029/2015 # misdated
+MF574587 # COL/PAN_00030/2015 # misdated
+KY785436 # BRA/2016/FC_DQ12D1 # large indel
+KY559010 # Brazil/2016/ZBRX8 # large indel
+KY559011 # Brazil/2016/ZBRX11 # large indel
+KX986761 # CX17 # large indel
+MF801405 # MEX/2016/mex27 # large indel
+MF801424 # MEX/2016/mex50 # large indel
+MF801377 # SLV/2016/ElSalvador_1055 # large indel
+VI20_12plex # USVI/20/2016 # large indel
 USVI/21/2016 # large indel
-USVI/23/2016 # large indel
-USVI/27/2016 # large indel
-USVI/30/2016 # large indel
-USVI/32/2016 # large indel
-Thailand/1605aTw # excess divergence
-VE_Ganxian # excess divergence
-ZK_YN001 # excess divergence
-Haiti/0029/2014 # contamination present
-Haiti/0033/2014 # contamination present
-Haiti/0036/2014 # contamination present
-Haiti/0054/2014 # contamination present
-Haiti/0074/2014 # contamination present
-Haiti/0097/2014 # contamination present
-mosquito/Haiti/1682/2016 # contamination present
+VI23_12plex # USVI/23/2016 # large indel
+VI27_1d # USVI/27/2016 # large indel
+VI30_1d # USVI/30/2016 # large indel
+VI32_12plex # USVI/32/2016 # large indel
+KY126351 # Thailand/1605aTw # excess divergence
+KU744693 # VE_Ganxian # excess divergence
+KY328290 # ZK_YN001 # excess divergence
+KY415986 # Haiti/0029/2014 # contamination present
+KY415987 # Haiti/0033/2014 # contamination present
+KY415990 # Haiti/0036/2014 # contamination present
+KY415988 # Haiti/0054/2014 # contamination present
+KY415989 # Haiti/0074/2014 # contamination present
+KY415991 # Haiti/0097/2014 # contamination present
+MF384325 # mosquito/Haiti/1682/2016 # contamination present
 ZF36_36S # contamination present
-MR766 # lab strain
-Aedes_sp/MEX_I_44/2016 # duplicate of Aedes_aegypti/MEX/MEX_I_44/2016
-Puerto_Rico/2015/PRVABC59 # duplicate of PRVABC59
-V15555 # highly diverged
-DK # lab strain
-DK23 # lab strain
-rGZ02a/2018 # highly diverged
-rGZ02p/2018 # highly diverged
-V211784 # highly diverged
-LMM/AG5643
-Faranah/18
+MK105975 # MR766 # lab strain
+KX856011 # Aedes_sp/MEX_I_44/2016 # duplicate of Aedes_aegypti/MEX/MEX_I_44/2016
+MK028857 # Puerto_Rico/2015/PRVABC59 # duplicate of PRVABC59
+MN025403 # V15555 # highly diverged
+MT505349 # DK # lab strain
+MT505350 # DK23 # lab strain
+MW680969 # rGZ02a/2018 # highly diverged
+MW680970 # rGZ02p/2018 # highly diverged
+OK054351 # V211784 # highly diverged
+MT478034 # LMM/AG5643
+OL414716 # Faranah/18