adding-a-claim

SOP: adding a claim to the wiki

Every biological claim must be (1) cited or (2) flagged as unsourced. This SOP covers the citation format and gap-marking conventions used everywhere on the wiki.

The claim → citation pattern

Fisetin reduced p16+ senescent cells by ~50% in aged C57BL/6 mice at 100 mg/kg
oral [^yousefzadeh2018]. **Not yet replicated in humans** #gap/needs-human-replication.
 
[^yousefzadeh2018]: [[studies/yousefzadeh-2018-fisetin-senolytic]] · n=20/group · randomized · p<0.01 · model: aged C57BL/6 mice

Three things happen in that paragraph:

1. The claim itself — quantitative, specific, includes route and dose.
2. A footnote with the evidence quality signal (n, design, p, organism) so a reader can judge the claim without chasing the link.
3. A #gap/... tag flagging that this is not yet human evidence.

Footnote field order (always)

[[studies/...]] · n=N · design · p<X · model: descriptor

Allowed design values:

• rct — randomized controlled trial
• observational (also cohort or case-control if you want specificity)
• meta-analysis
• systematic-review
• in-vivo — non-human animal study
• in-vitro — cell culture
• in-silico — computational only
• mendelian-randomization
• review — narrative review (low evidence weight on its own)

Omit p<X if not applicable (descriptive studies, meta-analyses with effect sizes, etc.).

When to use a bare DOI footnote vs a study page

• Study page + footnote: when the paper is the sole source for a claim, OR when you’re extracting multiple claims from it.
• Bare DOI footnote: when the claim is well-established and you’re citing one of many corroborating sources. Format:

  [^smith2020]: doi:10.1038/example · review · in [[archive_search:archive search --doi 10.1038/example]]
  

Gap markers

Tag every claim that needs further evidence:

Tag	When to use
#gap/needs-human-replication	Claim from model organism, no human equivalent yet
#gap/needs-replication	Single-study claim, no independent confirmation
#gap/contradictory-evidence	Two or more sources disagree (also link both in footnotes)
#gap/no-mechanism	Effect observed; mechanism unknown or hypothesized
#gap/dose-response-unclear	Efficacy known, optimal dose not
#gap/long-term-unknown	Short-term study only
#gap/unsourced	Claim added without citation; needs source
#stub	Page exists with frontmatter but no real content

These flow into gaps/README.md via Dataview.

How to phrase claims

• Be quantitative: “reduced X by 50%” not “significantly reduced X.”
• Include relevant covariates: dose, route, duration, organism, age, sex.
• Distinguish observation from mechanism: “X correlates with Y” vs “X causes Y” — the latter requires intervention or genetic evidence.
• Hedge appropriately: “appears to,” “in mice,” “at supraphysiological doses” when these caveats apply.
• Quote effect sizes with their CIs when reported in the source.
• Flag null results as informative, not failures.

When sources disagree

Don’t pick a side. Document both:

Caloric restriction extends lifespan ~30% in C57BL/6 mice [^liang2018], but recent
larger studies in heterogeneous mouse stocks find effects nearer 5–15% [^itp2024],
suggesting prior estimates were inflated by genetic background effects.
#gap/contradictory-evidence
 
[^liang2018]: [[studies/liang-2018-cr-c57bl6]] · n=80 · randomized · p<0.001 · model: C57BL/6
[^itp2024]: [[studies/itp-2024-cr-heterogeneous-mice]] · n=240/sex · randomized · p<0.05 · model: 4-way het mice

Common mistakes

• Citing a review as if it were primary evidence. Reviews aggregate; cite the underlying primary papers when possible.
• Citing the abstract and missing the actual result in the paper body.
• Omitting the organism. “X reduces inflammation” is different from “X reduces inflammation in aged mice.”
• Quoting a press release / news article. Always go back to the paper.
• Inventing a citation that doesn’t exist. If unsure, tag #gap/unsourced and surface in the next lint pass.

See also

• extracting-evidence — how to read a paper and create a study page
• lint-pass — how unsourced claims and gaps surface in periodic review
• _extrapolation-guide — model-organism translation rubric