H-28 - The form-food-matrix of choline, not total choline content, determines TMAO generation
The authors’ central mechanistic claim, discriminating between two rival TMAO models. Uncertain (the microbiome step is inferred, not directly measured here).
Link to original
H-29 - Habitual whole-egg consumption does not meaningfully raise TMAO in people with normal renal function
A candidate answer to the harm-mechanism sub-question, notable for coming from the lab (Hazen/Tang) that built the TMAO-harm hypothesis.
Link to original
H-30 - Impaired renal function may modify the TMAO response to egg-choline intake
A for-whom candidate answer flagged by the authors as a limitation; the trial deliberately restricted to normal renal function, so this remains untested here.
Link to original
H-42 - Eating eggs (-=2 yolks) raises TMAO production, feeding egg choline into the TMAO pathway
Well-supported for postprandial flux; its extrapolation to chronic exposure under normal (non-low-choline) diets and to hard outcomes is what remains uncertain.
Link to original
H-43 - The egg-induced TMAO rise may be clinically benign because it is postprandial and unaccompanied by inflammatory-oxidative changes
The paper’s more contestable interpretive claim, cutting against the TMAO-harm thread. Uncertain: n=6, only two surrogate harm markers over 24 h, and no hard-outcome data.
Link to original
H-56 - Some other egg-TMAO relationship
Residual catch-all so the egg-TMAO prior sums to 1; names an escape beyond the listed magnitude/condition characterizations.
Link to original
Carving
The sub-question is whether habitual egg intake produces a harm-relevant TMAO elevation and under what conditions. The members are competing characterizations of the magnitude/conditions of any egg-driven rise: negligible because egg phosphatidylcholine-choline is poorly converted relative to free choline (H-28); no meaningful fasting/steady-state rise in people with normal renal function (H-29); a rise only where renal clearance is impaired (H-30); a real postprandial production rise as egg choline enters the pathway at >=2 yolks (H-42); or a rise that occurs but is clinically benign, being postprandial and unaccompanied by inflammatory-oxidative change (H-43). Read as the single best bottom-line, at most one holds; the task’s explicit poles H-42 (postprandial rise) and H-29 (fasting null) are the anchors, reconciled in reality by the postprandial-versus-fasting distinction that the members turn into competing summary verdicts. Residual H-56 covers a meaningful chronic steady-state elevation with clinical import. Its relevance is gated by HC-8 (whether TMAO is causal), noted in depends_on.
Prior
# HC-9 prior — whether habitual egg intake produces a harm-relevant TMAO elevation, and under what
# conditions. Members in HC-9.hypotheses order:
# [H-28 form/matrix governs (egg-PC poorly converted), H-29 no meaningful fasting rise in normal-renal,
# H-30 renal-impairment-only rise, H-42 eggs (>=2 yolks) raise TMAO production, H-43 rise is clinically
# benign, H-56 residual].
# Evidence split: nothing is marked used_for_prior. Every inbound edge is a case-specific trial result —
# the Wilcox mechanistic RCT (CG-27) and the Miller dose-response (CG-28) — all discriminating, all left
# for step 8. no_observation_arguments.py --cluster HC-9 returns none, so no argument moved a number.
# The prior rests on the mechanistic outside view only. The members overlap in reality (a postprandial
# rise, a fasting null, and benignity can all co-occur); the cluster forces the single best *summary
# verdict*, and these weights price which verdict is a priori most apt.
# H-42 anchor — "eggs (>=2 yolks) raise TMAO production." The mechanistic default: two yolks deliver
# ~250-300 mg choline, choline is an established TMA precursor, gut microbiota convert it, so a measurable
# (postprandial) rise is the naively expected consequence of eating a choline-dense food. The most apt
# default headline before any egg-specific trial is priced.
w_h42 = 1.0 # unit anchor: the mechanistic default "eggs feed the pathway"
# H-29 no meaningful fasting/steady-state rise in normal-renal people, odds vs H-42. What must hold: acute
# postprandial flux does not carry into fasting/steady-state levels because renal clearance and homeostasis
# buffer them. Reference class: habitual-intake biomarker questions where the chronic steady state is what
# bears on harm. Near-co-equal to H-42 (the two named poles of the carving); just below because the
# mechanistic default is that a choline load does move the marker at least acutely.
odds_h29_vs_h42 = 0.85 # the fasting-null pole, near-co-equal to the postprandial-rise anchor
w_h29 = odds_h29_vs_h42 * w_h42
# H-43 the rise occurs but is clinically benign, odds vs H-42. What must hold: a real postprandial
# excursion that saturates above ~4 yolks and moves no inflammatory/oxidative marker, so it does not
# translate to harm. Reference class: postprandial-metabolite excursions without downstream surrogate
# change. A strong contender for the true bottom line of the whole harm question, hence third; below the
# two anchors because it makes an extra (harm-irrelevance) claim.
odds_h43_vs_h42 = 0.70 # makes an added harm-irrelevance claim on top of a rise, so below the anchors
w_h43 = odds_h43_vs_h42 * w_h42
# H-28 chemical form / food-matrix (not total choline) governs TMA generation, egg-PC poorly converted,
# odds vs H-42. What must hold: phosphatidylcholine-bound choline is substantially less TMA-bioavailable
# than free choline, and form dominates amount. Reference class: choline-form bioavailability studies. More
# specific than the anchors (requires form to be the controlling variable), so below them.
odds_h28_vs_h42 = 0.55 # a stronger, more specific mechanistic claim than the anchors
w_h28 = odds_h28_vs_h42 * w_h42
# H-30 renal-impairment-only rise ("for whom"), odds vs H-42. What must hold: a normal-renal null but a
# rise where clearance is reduced, since TMAO is almost exclusively renally cleared (mechanistically well
# grounded). Reference class: renally-cleared-metabolite accumulation in CKD. Narrower than the others as
# THE single best overall summary because it speaks only to the impaired-renal minority, so below them.
odds_h30_vs_h42 = 0.40 # mechanistically sound but a subgroup-only summary of the whole question
w_h30 = odds_h30_vs_h42 * w_h42
# H-56 residual — a relationship none of the five express: a meaningful *chronic steady-state* TMAO
# elevation with clinical import in ordinary (normal-renal) people, or a route other than egg choline. Its
# own argued weight (rule 4): disfavoured because renal clearance and homeostasis are expected to buffer
# steady-state levels, but real (e.g. a high-conversion + slow-clearance subpopulation sustaining an
# elevation).
odds_other = 0.35 # ~0.35 : 1 vs the unit anchor; ~9% of the mass once normalised
w_h56 = odds_other
prior("HC-9", [w_h28, w_h29, w_h30, w_h42, w_h43, w_h56])