H-39 - Choline is an essential nutrient for humans when dietary methionine and folate are not in excess
H-39 — At the time a novel claim (choline was not classified essential in humans); now established (an Adequate Intake was set in 1998). Establishes that eggs’ high choline content is nutritionally consequential rather than incidental.
Link to original
H-48 - Dietary choline is not an essential nutrient
Constructed complement: the null pole no curated paper states, needed so the choline-essentiality question is a two-answer space against H-39. Favoured if de novo synthesis covers requirements in ordinarily-nourished people.
Link to original
Carving
The sub-question is whether dietary choline, of which eggs are a dense source, is an essential nutrient for humans. Two exhaustive answers: essential when methionine and folate are adequate but not in excess (H-39) or not essential because endogenous PEMT synthesis suffices (H-48, constructed). Analysis-failure mode: the depletion evidence came from a short choline-free regimen in selected men; if de novo synthesis covers ordinary requirements in normally-nourished people, the analysis would favour H-39 when H-48 holds - the failure favours the constructed null. Historically H-39 prevailed - an Adequate Intake was set in 1998 - which informs but does not fix the prior. exhaustive_by_construction is set; H-48 is minted as the complement.
Prior
# HC-14 prior — is dietary choline (dense in eggs) an essential nutrient for humans. Exhaustive by
# construction (exactly one holds; no residual, so rule 4 does not apply). Members in HC-14.hypotheses
# order: [H-39 essential when methionine/folate adequate-but-not-in-excess, H-48 not essential (PEMT
# synthesis suffices)].
# Evidence split: nothing is marked used_for_prior. The only inbound edges are the Zeisel choline-
# depletion trial (CG-33: plasma-choline fall, ALT rise, cholesterol fall) — the case-specific test the
# cluster turns on, left in full for step 8. no_observation_arguments.py --cluster HC-14 returns none. The
# prior rests on the outside view: the standing expert-consensus classification, discounted for its
# evidentiary tier and for the analysis-failure mode the carving flags (a short choline-free regimen in
# selected men favours the constructed null).
# H-39 anchor — choline is (conditionally) essential. Reference class: nutrients formally recognised as
# required by an expert body on controlled depletion-repletion evidence. The US IOM set an Adequate Intake
# for choline in 1998 (550 / 425 mg/d men/women) — a standing consensus that it is required; such
# classifications are seldom later reversed. Most confident member. What it omits: the requirement is
# conditional on methionine/folate not being in excess, and interindividual (sex, genotype) variation is
# large.
p_essential_consensus_holds = 0.82 # base rate that an expert-body essentiality judgment backed by depletion-repletion data survives
w_h39 = 1.0 # unit anchor: the established/consensus reading
# H-48 not essential, odds vs H-39. What would have to be true: endogenous PEMT synthesis plus usual
# methionine/folate covers requirements in ordinarily-nourished people, so a choline-free diet does not
# *reliably* produce deficiency. Reference class / footing: choline carries an Adequate Intake, not an RDA
# (data were insufficient for an EAR/RDA — a weaker evidentiary tier), and de novo synthesis genuinely
# covers requirements for part of the population (e.g. premenopausal women, estrogen-upregulated PEMT).
# The carving's analysis-failure mode (depletion shown only in selected men) further tilts toward this null.
# Clearly the minority reading, but with real support.
base_odds_h48 = (1 - p_essential_consensus_holds) / p_essential_consensus_holds # odds identity from the base rate, not an estimate
caveat_inflation_h48 = 1.3 # AI-not-RDA evidentiary tier + synthesis covers part of the population + failure-mode favours the null
w_h48 = base_odds_h48 * caveat_inflation_h48 * w_h39
prior("HC-14", [w_h39, w_h48])