Update dataset card: single sqlite file architecture

README.md

@@ -21,18 +21,72 @@ size_categories:
 
 Synonym databases for deterministic grading in [GenomicsBench](https://github.com/AfterQuery/GenomicsBench), a realism-first benchmark for AI coding agents in computational biology.
 
-GenomicsBench ships with dozens of prebuilt graders for highly specific biological outputs. These databases power the synonym-aware matching behind those graders, so that *TP53* and *tumor protein p53* are recognized as the same gene, *hsa-miR-21-5p* and *MIMAT0000076* as the same miRNA, and *E. coli* and *Escherichia coli* as the same organism.
+GenomicsBench ships with dozens of prebuilt graders for highly specific biological outputs. This dataset powers the synonym-aware matching behind those graders, so that *TP53* and *tumor protein p53* are recognized as the same gene, *hsa-miR-21-5p* and *MIMAT0000076* as the same miRNA, and *E. coli* and *Escherichia coli* as the same organism.
 
-## Contents
+## Format
 
-Five tab-separated files, one per biological database. Each is a flat lookup table mapping canonical identifiers to all known names and synonyms.
+One read-only SQLite file, `grader_databases.sqlite` (~22 GB), with five tables -- one per source database. Lookup columns are pre-lowercased at build time (full Unicode `str.lower()`) and covered by composite indexes, so runtime queries are direct indexed seeks and process memory is O(query result) rather than O(database size).
 
-| File | Columns | Rows | Source | Description |
+| Table | Columns | Rows | Source | Description |
 |---|---|---|---|---|
-| `ncbi_gene_grader_db.tsv` | `organism  gene_id  name` | 218M | [NCBI Gene](https://www.ncbi.nlm.nih.gov/gene/) | Gene symbol, synonym, description, Entrez GeneID, and Ensembl ID resolution scoped by organism |
-| `ncbi_taxonomy_grader_db.tsv` | `canonical_name  name` | 3.2M | [NCBI Taxonomy](https://www.ncbi.nlm.nih.gov/taxonomy) | Scientific name, common name, abbreviation, and synonym resolution for all taxa |
-| `hmdb_grader_db.tsv` | `accession  name` | 2.0M | [HMDB](https://hmdb.ca/) | Metabolite name, synonym, IUPAC name, and accession resolution |
-| `mirbase_grader_db.tsv` | `accession  name` | 185K | [miRBase](https://mirbase.org/) | miRNA precursor and mature name/accession resolution, including deprecated entries |
-| `card_grader_db.tsv` | `aro_accession  name` | 19K | [CARD](https://card.mcmaster.ca/) | Antimicrobial resistance gene name, synonym, and ARO accession resolution |
+| `ncbi_gene` | `organism_lower  name_lower  gene_id` | ~219M | [NCBI Gene](https://www.ncbi.nlm.nih.gov/gene/) | Gene symbol, synonym, description, Entrez GeneID, and Ensembl ID resolution scoped by organism |
+| `ncbi_taxonomy` | `name_lower  canonical_name` | ~3.2M | [NCBI Taxonomy](https://www.ncbi.nlm.nih.gov/taxonomy) | Scientific name, common name, abbreviation, and synonym resolution for all taxa |
+| `hmdb` | `name_lower  accession` | ~1.5M | [HMDB](https://hmdb.ca/) | Metabolite name, synonym, IUPAC name, and HMDB accession resolution |
+| `mirbase` | `name_lower  accession` | ~157K | [miRBase](https://mirbase.org/) | miRNA precursor and mature name/accession resolution, including deprecated entries folded into live replacements |
+| `card` | `name_lower  aro_accession` | ~19K | [CARD](https://card.mcmaster.ca/) | Antimicrobial resistance gene name, synonym, and ARO accession resolution |
 
-For information on the provenance, licensing, or the ETL process generating this dataset, you should visit the [GenomicsBench GitHub repository](https://github.com/AfterQuery/GenomicsBench).
+Three of the tables (`ncbi_taxonomy`, `hmdb`, `mirbase`) use `WITHOUT ROWID` with `PRIMARY KEY(name_lower)` and are built via `INSERT OR IGNORE` so the first occurrence of any lowercased name wins. The other two (`ncbi_gene`, `card`) are 1:N and allow multiple rows per lookup key -- `gene_match` and `amr_match` do set-intersection at query time.
+
+## Usage
+
+Install GenomicsBench and call the match functions directly; downloading and caching happens on first use.
+
+```python
+from genomicsbench.grading import (
+    gene_match,
+    taxonomy_match,
+    mirna_match,
+    metabolite_match,
+    amr_match,
+)
+
+gene_match("TP53", "tumor protein p53", organism="Homo sapiens")  # True
+taxonomy_match("E. coli", "Escherichia coli")                     # True
+mirna_match("hsa-miR-21-5p", "MIMAT0000076")                      # True
+metabolite_match("Dextrose", "D-Glucose")                         # True
+amr_match("mecA", "PBP2A")                                        # True
+```
+
+Each match function is a thin wrapper over one indexed SQL query plus a case-insensitive string-equality fallback (used when either side is absent from the database, so misses don't crash on out-of-distribution inputs). The public API signatures, return types, and fallback semantics are unchanged from the previous TSV-backed release -- consumers upgrade by bumping their `genomicsbench` package version.
+
+Each lookup module opens a thread-local read-only connection on first use with read-tuning PRAGMAs (`query_only=1`, `cache_size=-20000`, `temp_store=MEMORY`, `mmap_size=268435456`), so the dataset works correctly under multi-threaded consumers like the GenomicsBench async orchestrator. SQLite supports unlimited concurrent readers via shared file locks.
+
+## Direct access
+
+If you want to query the tables yourself without installing GenomicsBench, open the file read-only:
+
+```python
+import sqlite3
+from huggingface_hub import hf_hub_download
+
+path = hf_hub_download(
+    repo_id="AfterQuery/GenomicsBench-grader-databases",
+    filename="grader_databases.sqlite",
+    repo_type="dataset",
+)
+conn = sqlite3.connect(f"file:{path}?mode=ro", uri=True)
+
+rows = conn.execute(
+    "SELECT gene_id FROM ncbi_gene "
+    "WHERE organism_lower = ? AND name_lower = ?",
+    ("homo sapiens", "tp53"),
+).fetchall()
+```
+
+All lookup columns are pre-lowercased; lowercase your query inputs in Python before binding. The table schemas are stable across releases: any schema change is a breaking revision of this dataset.
+
+## Provenance and licensing
+
+The five source databases retain their upstream licenses. GenomicsBench packages them into a single SQLite file for efficient runtime access; see the [GenomicsBench GitHub repository](https://github.com/AfterQuery/GenomicsBench) for the full ETL pipeline under `scripts/databases/` and its step-by-step runbook in `scripts/databases/README.md`.
+
+Rebuilds ship as new revisions on this dataset whenever an upstream source publishes a new release.