Card: metadata v2.2 (tag metadata-v2.2-20261106, corrections from human review)

README.md

@@ -248,15 +248,15 @@ catalog has no per-sample license column. See the
 reuse terms and citation, and cite the original studies with the data you use (`sda-db cite` writes BibTeX for
 a selection).
 
-## Curated sample metadata v2 (2026-11-05)
+## Curated sample metadata v2 (version 2.2, 2026-11-06)
 
 `metadata/v2/` adds a curated table on top of the same 12,163 samples. No data file changed: the paper release is
-still tag `v1.0.0-20261025`.
+still tag `v1.0.0-20261025`; the current metadata is tag `metadata-v2.2-20261106`.
 
 - `samples_metadata_v2.parquet` / `.csv` has one row per sample:
   - organ (43 sites);
-  - UBERON tissue term (11,606 samples);
-  - MONDO disease terms (6,477);
+  - UBERON tissue term (11,605 samples);
+  - MONDO disease terms (6,479);
   - sample kind, disease status (normal / disease / perturbation / not stated), life stage and sex;
   - NCBI taxonomy, and the species implied by the gene IDs;
   - corrected platform family and assay name.
@@ -267,7 +267,7 @@ still tag `v1.0.0-20261025`.
   sample's GEO record, so it could not invent identifiers. Fixed rules handle unambiguous ages.
 - **Quality, estimated on 300 random samples.** An independent model judged each label and every "wrong" verdict
   was read by hand. Estimated wrong rates: organ 0.3%, tissue term 0.7%, sample kind 0.7%, life stage 1.7%,
-  disease 3.0%. Labels judged correct were not re-read, so these are estimates.
+  disease 3.0%. Labels judged correct were not re-read, so these are estimates. A human reader then reviewed 200 samples, and the errors found were corrected in version 2.2.
 - **Known limit.** A study's disease sometimes lands on a control whose own record does not say it is a control
   (about 1.5% of samples). Use `annotation_low_confidence`, `metadata_flags` and the raw fields when this matters.
 - Full description: `metadata/v2/README.md`.