Update to use koza split command to produce separate node files for e…

…ach included taxon
monarch-initiative · Oct 5, 2024 · 77547dd · 77547dd
1 parent 1a19b7e
commit 77547dd
Show file tree

Hide file tree

Showing 4 changed files with 11 additions and 7 deletions.
diff --git a/Makefile b/Makefile
@@ -74,6 +74,7 @@ download:
 .PHONY: run
 run: download
 	$(RUN) ingest transform
+	$(RUN) koza split output/ncbi_gene_nodes.tsv in_taxon --remove-prefixes --output-dir output/by_taxon
 	$(RUN) python scripts/generate-report.py
 
 

diff --git a/poetry.lock b/poetry.lock
diff --git a/pyproject.toml b/pyproject.toml
@@ -9,7 +9,7 @@ readme = "README.md"
 [tool.poetry.dependencies]
 python = "^3.10"
 importlib-metadata = ">=4.8.0"
-koza = ">=0.6.0"
+koza = "0.7.1"
 kghub-downloader = ">=0.3.8"
 kgx = ">=2.4.0"
 biolink-model = "^4.2.0"

diff --git a/src/ncbi_gene/transform.yaml b/src/ncbi_gene/transform.yaml
@@ -44,10 +44,13 @@ columns:
 
 header: 0
 
+# This likely will need adjusting based on the included taxa below
+min_node_count: 2500000
+
 filters:
   - inclusion: "include"
     column: "tax_id"
-    filter_code: "in"
+    filter_code: "in_exact"
     value:
       - "10036"
       - "10042"