Add CLI normalizer

Dockerfile

@@ -27,6 +27,7 @@ RUN apt-get install -y screen
 RUN apt-get install -y vim
 RUN apt-get install -y rsync
 RUN apt-get install -y jq
+RUN apt-get install -y duckdb
 
 # Create a non-root-user.
 RUN adduser --home ${ROOT} --uid 1000 nru

Snakefile

@@ -20,6 +20,11 @@ include: "src/snakefiles/duckdb.snakefile"
 include: "src/snakefiles/reports.snakefile"
 include: "src/snakefiles/exports.snakefile"
 
+# Some global settings.
+import os
+os.environ['TMPDIR'] = config['tmp_directory']
+
+# Top-level rules.
 rule all:
     input:
         # See rule all_outputs later in this file for how we generate all the outputs.

config.yaml

@@ -1,15 +1,28 @@
+# Overall inputs and outputs.
 input_directory: input_data
 download_directory: babel_downloads
 intermediate_directory: babel_outputs/intermediate
 output_directory: babel_outputs
+tmp_directory: babel_downloads/tmp
 
+# Versions that need to be updated on every release.
 biolink_version: "4.2.6-rc5"
-umls_version: "2024AB"
-rxnorm_version: "03032025"
+umls_version: "2025AA"
+rxnorm_version: "07072025"
 drugbank_version: "5-1-13"
 
+#
+# PROTEINS
+#
+
+# Chris Bizon prepared a list of UMLS/UniProtKB mappings which we download and use.
 UMLS_UniProtKB_download_raw_url: "https://raw.githubusercontent.com/cbizon/UMLS_UniProtKB/refs/heads/main/outputs/UMLS_UniProtKB.tsv"
 
+#
+# The rest of these configs need to be cleaned up.
+#
+
+
 ncbi_files:
   - gene2ensembl.gz
   - gene_info.gz
@@ -145,6 +158,7 @@ protein_concords:
   - PR
   - NCIT_UniProtKB
   - NCIT_UMLS
+  - UMLS
   - UMLS_UniProtKB
 
 protein_outputs:
@@ -282,6 +296,9 @@ chemical_outputs:
 drugchemicalconflated_synonym_outputs:
   - DrugChemicalConflated.txt
 
+geneproteinconflated_synonym_outputs:
+  - GeneProteinConflated.txt
+
 taxon_labels:
   - NCBITaxon
   - MESH

requirements.lock

@@ -92,7 +92,7 @@ pronto==2.7.0
 propcache==0.3.1
 psutil==7.0.0
 psycopg2-binary==2.9.10
-PuLP==3.1.1
+PuLP==2.7.0
 pydantic==2.11.4
 pydantic_core==2.33.2
 PyJSG==0.11.10
@@ -128,7 +128,7 @@ ShExJSG==0.8.2
 six==1.17.0
 smart-open==7.1.0
 smmap==5.0.2
-snakemake==9.3.3
+snakemake==7.32.4
 snakemake-interface-common==1.17.4
 snakemake-interface-executor-plugins==9.3.5
 snakemake-interface-logger-plugins==1.2.3
@@ -142,10 +142,12 @@ SQLAlchemy==2.0.40
 SQLAlchemy-Utils==0.38.3
 sssom==0.4.15
 sssom-schema==1.0.0
+stopit==1.1.2
 stringcase==1.2.0
 tabulate==0.9.0
 tenacity==8.5.0
 throttler==1.2.2
+toposort==1.10
 tqdm==4.67.1
 traitlets==5.14.3
 types-python-dateutil==2.9.0.20241206

requirements.txt

@@ -31,3 +31,8 @@ duckdb
 # on checking if it's online via http://httpstat.us/200, which is often offline. My branch of this
 # https://github.com/gaurav/apybiomart/tree/change-check-url and changes that to https://example.org.
 git+https://github.com/gaurav/apybiomart.git@change-check-url
+
+# Added by Gaurav, Aug 2025 to check for memory information while Babel is running.
+psutil
+humanfriendly
+py-spy