feat: replace OrphaPacket by ORDO CSV (#84)

Snakefile

@@ -86,7 +86,7 @@ rule all:
         #
         # genes
         f"work/download/genes/rcnv/2022/Collins_rCNV_2022.dosage_sensitivity_scores.tsv.gz",
-        f"work/download/genes/orphapacket/{DV.orphapacket}/orphapacket.tar.gz",
+        f"work/download/genes/ordo/{DV.ordo}/ordo.csv",
         "work/download/genes/alphamissense/1/AlphaMissense_gene_hg38.tsv.gz",
         f"work/genes/dbnsfp/{DV.dbnsfp}/genes.tsv.gz",
         "work/genes/decipher/v3/decipher_hi_prediction.tsv.gz",
@@ -96,7 +96,7 @@ rule all:
         f"work/genes/gnomad/{DV.gnomad_constraints}/gnomad_constraints.tsv",
         f"work/genes/hgnc/{DV.today}/hgnc_info.jsonl",
         f"work/genes/omim/{DV.hpo}+{DV.today}/omim_diseases.tsv",
-        f"work/genes/orphapacket/{DV.orphapacket}+{DV.today}/orpha_diseases.tsv",
+        f"work/genes/ordo/{DV.ordo}/orpha_diseases.tsv",
         "work/genes/rcnv/2022/rcnv_collins_2022.tsv",
         "work/genes/shet/2019/shet_weghorn_2019.tsv",
         f"work/genes/clingen/{DV.today}/ClinGen_gene_curation_list_GRCh37.tsv",
@@ -177,7 +177,7 @@ rule all:
         f"output/full/annonars/cons-grch37-{DV.ucsc_cons_37}+{PV.annonars}/rocksdb/IDENTITY",
         f"output/full/annonars/cons-grch38-{DV.ucsc_cons_38}+{PV.annonars}/rocksdb/IDENTITY",
         # ----- genes
-        f"output/full/annonars/genes-{DV.acmg_sf}+{DV.gnomad_constraints}+{DV.dbnsfp}+{DV.hpo}+{DV.orphapacket}+{DV.today}+{PV.annonars}/rocksdb/IDENTITY",
+        f"output/full/annonars/genes-{DV.acmg_sf}+{DV.gnomad_constraints}+{DV.dbnsfp}+{DV.hpo}+{DV.ordo}+{DV.today}+{PV.annonars}/rocksdb/IDENTITY",
         # -- worker data
         f"output/full/worker/genes-regions-grch37-{DV.refseq_37}+{PV.worker}/refseq_genes.bin",
         f"output/full/worker/genes-regions-grch37-{DV.ensembl_37}+{PV.worker}/ensembl_genes.bin",
@@ -350,7 +350,7 @@ include: "rules/work/genes/mehari_data_tx.smk"
 include: "rules/work/genes/ncbi.smk"
 include: "rules/work/genes/omim.smk"
 include: "rules/work/genes/panelapp.smk"
-include: "rules/work/genes/orphapacket.smk"
+include: "rules/work/genes/ordo.smk"
 include: "rules/work/genes/rcnv.smk"
 include: "rules/work/genes/shet.smk"
 include: "rules/work/genes/domino.smk"

download_urls.yml

@@ -101,7 +101,7 @@
     strategy: manual
     count: null
 
-- url: https://github.com/Orphanet/orphapacket/archive/refs/tags/v10.1.tar.gz
+- url: https://data.bioontology.org/ontologies/ORDO/download?apikey=8b5b7825-538d-40e0-9e9e-5ab9274a9aeb&download_format=csv
   excerpt_strategy:
     strategy: no-excerpt
     count: null

rules/output/annonars/genes.smk

@@ -11,7 +11,7 @@ rule output_annonars_genes:  # -- build annonars genes RocksDB file
         hgnc="work/genes/hgnc/{date}/hgnc_info.jsonl",
         ncbi="work/genes/entrez/{date}/gene_info.jsonl",
         omim="work/genes/omim/{v_hpo}+{date}/omim_diseases.tsv",
-        orpha="work/genes/orphapacket/{v_orpha}+{date}/orpha_diseases.tsv",
+        orpha="work/genes/ordo/{v_orpha}/orpha_diseases.tsv",
         panelapp="work/download/genes/panelapp/{date}/panelapp.jsonl",
         rcnv="work/genes/rcnv/2022/rcnv_collins_2022.tsv",
         shet="work/genes/shet/2019/shet_weghorn_2019.tsv",
@@ -65,6 +65,6 @@ rule output_annonars_genes:  # -- build annonars genes RocksDB file
             \
             --value v_annonars={wildcards.v_annonars} \
             --value v_downloader={PV.downloader} \
-            --value v_orphapacket={wildcards.v_orpha} \
+            --value v_ordo={wildcards.v_orpha} \
         > {output.spec_yaml}
         """

rules/output/annonars/genes.spec.yaml

@@ -38,7 +38,7 @@ x-created-from:
   - name: OMIM
     version: {{ today }}
   - name: ORDO
-    version: {{ v_orphapacket }}
+    version: {{ v_ordo }}
   - name: rCNV pHaplo/pTriplo scores
     version: 2022-Collins-et-al
   - name: sHet scores

rules/work/genes/ordo.smk

@@ -0,0 +1,34 @@
+## Rules related to ORDO download
+
+
+rule genes_ordo_download:  # -- download ORDO CSV file
+    output:
+        csv=f"work/download/genes/ordo/{DV.ordo}/ordo.csv",
+        csv_md5=f"work/download/genes/ordo/{DV.ordo}/ordo.csv.md5",
+    shell:
+        r"""
+        wget --no-check-certificate \
+            -O {output.csv} \
+            'https://data.bioontology.org/ontologies/ORDO/submissions/28/download?apikey=8b5b7825-538d-40e0-9e9e-5ab9274a9aeb&download_format=csv'
+
+        md5sum {output.csv} > {output.csv_md5}
+        """
+
+rule genes_ordo_convert:  # -- postprocess file for HGNC gene IDs
+    input:
+        csv="work/download/genes/ordo/{version}/ordo.csv",
+    output:
+        tsv="work/genes/ordo/{version}/orpha_diseases.tsv",
+        tsv_md5="work/genes/ordo/{version}/orpha_diseases.tsv.md5",
+    shell:
+        """
+        export TMPDIR=$(mktemp -d)
+        trap "rm -rf $TMPDIR" ERR EXIT
+
+        python ./scripts/genes-orpha-diseases.py {input.csv} \
+        | qsv sort -d '\t' \
+        | qsv fmt -t '\t' \
+        > {output.tsv}
+
+        md5sum {output.tsv} > {output.tsv}.md5
+        """

scripts/genes-orpha-diseases.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python
-"""Helper script to extract gene-disease association from orphapacket."""
+"""Helper script to extract gene-disease association from ORDO CSV."""
 
 import csv
 import json
@@ -8,13 +8,13 @@
 
 
 def main():
-    symbol_to_hgnc = {}
+    records = {}
     with open(sys.argv[1], "rt") as inputf:
-        reader = csv.DictReader(inputf, delimiter="\t")
+        reader = csv.DictReader(inputf, delimiter=",")
         for record in reader:
-            symbol_to_hgnc[record["gene_symbol"]] = record["hgnc_id"]
+            records[record["gene_symbol"]] = record["hgnc_id"]
 
-    print(f"# xlink entries: {len(symbol_to_hgnc)}", file=sys.stderr)
+    print(f"# xlink entries: {len(records)}", file=sys.stderr)
 
     base_path = pathlib.Path(sys.argv[2])
     print("\t".join(["hgnc_id", "orpha_id", "disease_name"]))

varfish_db_downloader/versions.py

@@ -102,8 +102,8 @@ class DataVersions:
     acmg_sf: str
     #: HPO
     hpo: str
-    #: OrphaPacket
-    orphapacket: str
+    #: ORDO
+    ordo: str
     #: Pathogenic MMS
     patho_mms: str
     #: Mehari transcript data.
@@ -162,7 +162,7 @@ class DataVersions:
     dbsnp="b151",
     acmg_sf="3.1",
     hpo="20230606",
-    orphapacket="10.1",
+    ordo="4.4",
     patho_mms="20220730",
     mehari_tx="0.4.4",
     clinvar_release=CLINVAR_RELEASE,