Skip to content

GenBank flat-file record card on the Sequence Detail page

Motivation

A researcher compared the dashboard's Sequence Detail page (/sequence/:accession) with the public NCBI nuccore page for accession OZ254605.1 and noted that NCBI leads with the canonical GenBank flat-file header block (LOCUS / DEFINITION / ACCESSION / VERSION / DBLINK / KEYWORDS / SOURCE / ORGANISM), while the dashboard only showed the same data scattered across structured cards. They asked whether the dashboard could reproduce that familiar layout.

User-facing change

  • The Sequence Detail page now renders a "GenBank record" card directly under the header summary. It reproduces the NCBI flat-file header block verbatim in a monospace block with the classic 12-column tag field, 79-column line width, and continuation lines indented to column 13.
  • The block is composed from data the page already fetches via getNuccoreGenBank (LOCUS line from locus/length/moltype/topology/division/update_date, wrapped DEFINITION, ACCESSION, VERSION, DBLINK from xrefs, KEYWORDS, SOURCE, and the ORGANISM sub-keyword with the wrapped taxonomy lineage).
  • The existing structured cards (Sample & source, Taxonomy, Features, References, FASTA preview) are unchanged — the flat-file card is purely additive.

API / IaC diff summary

  • Backend (api/services/ncbi/nuccore.py): added _parse_keywords to parse the GBSeq_keywords → GBKeyword list (cap 32, each truncated to 120 chars) and a new keywords: list[str] field in the fetch_nuccore_genbank return dict. This was the only GenBank header field not already exposed, so KEYWORDS now renders faithfully (. when empty, matching the GenBank convention). No route/model change — the /ncbi/nuccore/{accession}/genbank route returns the dict directly.
  • Frontend types (web/src/api/ncbi.ts): added keywords: string[] to the NuccoreGenBank interface.
  • Frontend page (web/src/pages/sequence/SequenceDetail.tsx): added module-level genbankTag / genbankWrap / genbankFlatLines helpers, a flatRecord memo, and the "GenBank record" glass card.
  • No IaC change.

Validation evidence

  • uv run pytest -q api/tests/test_ncbi_nuccore.py → 52 passed (includes the updated test_fetch_nuccore_genbank_parses_record asserting keywords == ["RefSeq", "MANE Select"] against the extended GBSeq_keywords fixture).
  • uv run ruff check api/services/ncbi/nuccore.py api/tests/test_ncbi_nuccore.py → clean.
  • cd web && npm run build → built successfully (SequenceDetail-*.js emitted).