# Incoprea Research supplementary working library

SOLE intake: `/Users/tobymikle/Desktop/incoprea-research-publish`

Planned public root: https://incoprea.com/research-library/v2/

Current public article: https://incoprea.com/research/pharmacogenomics-coverage/

This fresh package was built only from the designated Desktop intake. No earlier staging folder or ZIP was used as content. The submitted source is frozen byte-for-byte under the local source-freeze directory, outside this public ZIP. The intake files were not edited. This is a publish-first working library, not an independently validated scientific release. All existing supplementary pages, including formerly held pages, are retained with visible working-draft labels. Internal contents and submenus are retained; no global Incoprea header is added.

## Correction ledger

1. A missing API field does not mean that medical evidence is absent; this is an audit of selected database fields.
2. support_n is an allele-record sample-support proxy, not a phenotype denominator, unique-person count, or effective sample size.
3. Sampling-noise and artefact conclusions are unsupported: this audit did not reconstruct the required denominators or propagate uncertainty.
4. The composite CPIC high-priority proportion measures a clinical-attention flag; it is not harm, adverse-reaction risk, or disease prevalence.
5. Older research-priority rankings and their figure/table are unvalidated. They must not be used as clinical or research-policy recommendations.
6. Manuscript HTML, Markdown and DOCX are superseded v1 exports despite their inclusion in a newer intake bundle. Their scientific text is preserved. HTML links identify them as superseded drafts / do not cite, and manuscript HTML has a prominent banner.
7. Exploratory notes are available as working drafts. Their submitted status labels do not assert independent validation. Their raw payloads are omitted; source endpoints and query descriptions remain in the notes.
8. Intake manifests and revision/status documents are preserved under intake-records/. Their historical status claims are source records, subject to this working-library ledger.

## Sources and reproducibility

The submitted CPIC analysis describes a 30 October 2026 snapshot from https://api.cpicpgx.org/v1/. The raw snapshot is intentionally not redistributed. A later live fetch is not necessarily the same historical input. Code, processed CSV values and figures are preserved rather than scientifically reworked.

Exploratory notes identify ClinicalTrials.gov API v2 and NCBI E-utilities (PubMed / ClinVar). Consult each note’s dates, query scope and limitations. No data retrieval or new scientific analysis was performed while packaging.

SOURCE-PROVENANCE.json records every included intake path, exact source hash, public path, package hash, generated page, and exclusion. Its own hash is omitted to avoid self-reference.

## Exclusions

- `notes/clinical-trials-landscape/raw.json` — raw third-party snapshot excluded
- `notes/clinvar-pgx-genes/raw.json` — raw third-party snapshot excluded
- `notes/publication-attention/raw.json` — raw third-party snapshot excluded
- `site/_archive/HANDOFF.md` — archived intake material excluded
- `site/_archive/PUBLISH-ORDER.md` — archived intake material excluded

Archived handoffs, raw snapshots, hidden/configuration files, symlinks and detected credential material are excluded. Unreviewed material is not excluded merely because it is unreviewed.

## Deployment

Extract ZIP contents directly into research-library/v2/ under the established Incoprea web root. The ZIP has no enclosing folder. library-index.html is the complete page/data/source inventory. Publishing this working library does not validate its scientific conclusions or replace the current article.