Home / About
About
A navigable reading of ten 2026 papers on NLR immune receptor biology. Everything here is generated from a structured data layer; nothing is written directly into HTML.
How to read it
Enter from any angle: a topic, a receptor family, a species, a method, an effector, or a paper. Every claim on a paper page carries a source locator, a figure, table or database accession, so any statement can be traced back to where it came from.
Rules the content follows
- No statement without a source. Anything untraceable is marked unverified and excluded.
- No invented citations, accessions or gene names.
- Papers are identified by title and ID. The site is organised by science, not by who did the work.
- Where the papers disagree with each other or with prior literature, the disagreement is stated rather than smoothed.
How it was built
- Ten PDFs extracted into one frozen schema, one file per paper, 101 findings each with a locator.
- Entities reconciled into a graph; every merge and every refused merge logged.
- Reference lists for all ten papers retrieved from OpenAlex: 625 citation edges, 424 distinct works.
- Prose written from the JSON, then this site generated from both.
Sources
- Data layer:
data/— manifest, per-paper extractions, entities, taxonomy, citation graph - Build:
scripts/build_entities.py,scripts/build_site.py - Decisions and caveats:
notes/, includingsource-errors.mdandalias-decisions.md