Rebuild CAB external temporal and failure union outputs

README_external_rebuild.md

@@ -0,0 +1,27 @@
+# CAB external rebuild scaffold
+
+This package adds scripts/configuration for raw external rebuild work.
+
+It does not restore deprecated CAB claims automatically. It writes manifests, checksums, logs, parsed tables, failures, and publication-safe claim status files.
+
+Install dependency:
+
+```bash
+python -m pip install pyyaml
+```
+
+Run on Linux/macOS/Git Bash:
+
+```bash
+bash run_external_rebuild.sh
+```
+
+Run on Windows PowerShell:
+
+```powershell
+powershell -ExecutionPolicy Bypass -File .\run_external_rebuild.ps1
+```
+
+gnomAD requires a local extracted sites table or custom API-backed lookup because full v4 resources are large. Set `gnomad.local_sites_table` in `config/external_resources.yaml`.
+
+Patch-clamp supplementary tables require manual source/license review. No copyrighted article text scraping is performed.

config/external_resources.yaml

@@ -0,0 +1,107 @@
+# CAB external rebuild resource configuration.
+# Public endpoints were identified from official/public documentation where possible.
+# Verify exact archived ClinVar snapshot file names before publication.
+# Do not convert failed downloads into publication claims. Yes, apparently that needs to be written down.
+
+clinvar:
+  current_variant_summary:
+    resource_name: ClinVar current variant_summary
+    version: current
+    release_date: current
+    url: https://ftp.ncbi.nlm.nih.gov/pub/clinvar/tab_delimited/variant_summary.txt.gz
+    output: data/raw/clinvar/current/variant_summary.txt.gz
+    license_or_terms: NCBI ClinVar public FTP; verify current terms at access time.
+  current_submission_summary:
+    resource_name: ClinVar current submission_summary
+    version: current
+    release_date: current
+    url: https://ftp.ncbi.nlm.nih.gov/pub/clinvar/tab_delimited/submission_summary.txt.gz
+    output: data/raw/clinvar/current/submission_summary.txt.gz
+    license_or_terms: NCBI ClinVar public FTP; verify current terms at access time.
+  baseline_variant_summary:
+    resource_name: ClinVar baseline variant_summary
+    version: 2023-01
+    release_date: 2023-01
+    url: https://ftp.ncbi.nlm.nih.gov/pub/clinvar/tab_delimited/archive/variant_summary_2023-01.txt.gz
+    output: data/raw/clinvar/2023-01/variant_summary_2023-01.txt.gz
+    license_or_terms: NCBI ClinVar public FTP; verify current terms at access time.
+  followup_variant_summary:
+    resource_name: ClinVar follow-up variant_summary
+    version: 2026-04
+    release_date: 2026-04
+    url: https://ftp.ncbi.nlm.nih.gov/pub/clinvar/tab_delimited/archive/variant_summary_2026-04.txt.gz
+    output: data/raw/clinvar/2026-04/variant_summary_2026-04.txt.gz
+    license_or_terms: NCBI ClinVar public FTP; verify current terms at access time.
+
+clingen:
+  gene_disease_validity:
+    resource_name: ClinGen Gene-Disease Validity
+    version: current
+    release_date: current
+    # If this direct CSV endpoint changes, download from https://search.clinicalgenome.org/kb/downloads and update this URL.
+    url: https://search.clinicalgenome.org/kb/gene-validity/download
+    output: data/raw/clingen/current/gene_disease_validity.csv
+    license_or_terms: ClinGen public resource; verify terms at access time.
+  cspec_registry:
+    resource_name: ClinGen CSpec Registry landing page
+    version: current
+    release_date: current
+    url: https://cspec.genome.network/cspec/ui/svi/
+    output: data/raw/clingen/current/cspec_registry_landing.html
+    license_or_terms: ClinGen CSpec public pages/API; structured export may require manual endpoint discovery.
+  evidence_repository:
+    resource_name: ClinGen Evidence Repository landing page
+    version: current
+    release_date: current
+    url: https://erepo.clinicalgenome.org/evrepo/
+    output: data/raw/clingen/current/evidence_repository_landing.html
+    license_or_terms: ClinGen Evidence Repository; structured export may require API/manual retrieval.
+
+gnomad:
+  release: v4.1.1_or_latest_v4
+  documentation_url: https://gnomad.broadinstitute.org/downloads
+  local_sites_table: ""
+  required_columns:
+    - chr
+    - pos
+    - ref
+    - alt
+    - global_af
+    - popmax_af
+    - popmax_population
+    - ac
+    - an
+
+mavedb:
+  api_base_url: https://api.mavedb.org/api/v1
+  genes:
+    - KCNQ1
+    - KCNH2
+    - KCNE1
+    - SCN5A
+    - RYR2
+    - CACNA1C
+    - CASQ2
+    - HCN4
+    - ANK2
+    - CALM1
+    - CALM2
+    - CALM3
+    - TRDN
+
+patch_clamp:
+  manual_inventory_csv: data/raw/patch_clamp/manual_source_inventory.csv
+  priority_genes:
+    - KCNQ1
+    - KCNH2
+    - SCN5A
+    - RYR2
+    - CACNA1C
+    - KCNE1
+    - CASQ2
+
+cab_inputs:
+  master_candidates:
+    - data/processed/cab_causal_architecture_assignments.csv
+    - data/processed/cab_routing_assignments.csv
+  failure_membership: data/processed/cab_failure_membership_explicit.csv