Add clinical_supplement struct column to patient rows (BCR biotab data: patient/follow_ups/ntes/drugs/radiations/ablations/omfs sub-fields, per-project schema). Patients view now surfaces the same BCR data the tabular view exposes.

Card restructure: symmetric two-views design (no primary/secondary), shared Data Model + How-Built combined, forward-compatible 'a supplement' phrasing for survival_derived, trimmed disclaimer.

Card cleanup: trim discovery-detail prose, drop project lists + misleading source line, share biospecimen tree + survival section across both cards, reorder so survival_derived sits after the GDC-source story, add non-affiliation note to disclaimer.

Schema cleanup: drop cdr_* + flat survival cols, add survival_derived struct (Liu et al. 2018 algorithm). Cards now share a single biospecimen-tree source of truth + new Survival endpoints section.

Rebuild dfi_event/dfi_time with BCR biotab Clinical Supplement integration (97% reduction in under-population vs Liu's CDR; agreement rate 77.2% -> 90.1% with 30-day time tolerance)

Card: consolidated-specific GDC requests wording (nested arrays, not flat tables); drop prototype disclaimer

Card: drop private-dataset caveats — both repos are now public

Full molecular for all 33 TCGA projects (11,428 patients)

Add full molecular for batch 2: PCPG/ESCA/SARC/CESC/KIRP/GBM/LIHC/OV/BLCA/STAD

Add full molecular for batch 1: UCS/ACC/UVM/MESO/KICH/THYM/LAML/TGCT/READ/PAAD

Add clinical for all 33 TCGA projects (11,428 patients total)

Card: simplify How this dataset is built section; link to source for full payloads

Trim FILE_FIELDS to only what we consume; document /files default behavior

Tighten /files filters: lock data_format + experimental_strategy + analysis.workflow_type per modality

Card: add How this dataset is built section with verbatim GDC request payloads

Rename per-project files train.parquet → data.parquet

drop unsupported TCGA library-prep claim; cite GDC pipeline harmonization quote instead

row groups + page index for HF Data Studio; gdc_portal_url; strict pydantic models; deduplicated dataset card

Add TCGA-LUAD (585 patients) + full GDC dictionary-derived schema + TcgaHfPatient pydantic reference impl + dictionary version capture

Preserve full GDC biospecimen hierarchy (samples/portions/analytes/aliquots); add TcgaHfPatient pydantic reference impl

Add gene expression (CHOL+DLBC) and mutations (CHOL+DLBC) to patient rows

Capture GDC data release version (45.0) in card and per-project gdc_status.json

Deterministic sort: diagnoses/treatments/follow_ups by (days_to_*, *_id); exposures/family_histories by id

Slim down dataset card to stable bits (license + redistribution)

Switch to per-project HF subsets (one config per TCGA-* project)

Update TCGA patients (open access) dataset