What the analysis says
The cluster asks whether habitual egg intake produces a harm-relevant TMAO elevation, and under what conditions — members as competing single-best summary verdicts: H-28 - The form-food-matrix of choline, not total choline content, determines TMAO generation, H-29 - Habitual whole-egg consumption does not meaningfully raise TMAO in people with normal renal function, H-30 - Impaired renal function may modify the TMAO response to egg-choline intake, H-42 - Eating eggs (-=2 yolks) raises TMAO production, feeding egg choline into the TMAO pathway, H-43 - The egg-induced TMAO rise may be clinically benign because it is postprandial and unaccompanied by inflammatory-oxidative changes, and residual H-56 - Some other egg-TMAO relationship. The mechanistic-outside-view prior anchored on H-42 (a choline-dense food should feed the pathway): [0.143, 0.221, 0.104, 0.260, 0.182, 0.091]. Posterior [0.123, 0.271, 0.077, 0.221, 0.253, 0.055]: no-fasting-rise H-29 rose to lead (0.271) and benign H-43 rose to a near-tie (0.182 to 0.253), while the postprandial-rise anchor H-42 fell (0.260 to 0.221). Two RCTs pull opposite anchors: CG-27 - HC-9 joint over O-62+O-63+O-64+O-65+O-67 (Wilcox 2021: 4 eggs/day x4wk left fasting TMAO flat while free choline raised it 6-fold and phosphatidylcholine did not) anchored on H-28 but its egg fasting-null strongly favoured H-29 (0.85x) and cut H-42 (0.35x); CG-28 - HC-9 joint over O-103+O-104+O-105+O-106+O-107 (Miller n=6: egg TMAO rises dose-dependently but acute-postprandial, plateaus 4-6 yolks, no hsCRP/oxLDL change) anchored on H-43 (1.0) and backed H-42 (0.9x). The crux is postprandial-versus-fasting, reconciled by O-105 (the rise sits off the fasting axis) — which is exactly why H-29 and H-43 co-lead.
What the model may not capture
The members are competing summary verdicts over phenomena that co-occur in reality — a postprandial rise, a fasting null, and clinical benignity can all be simultaneously true — so forcing one bottom-line is partly artificial, and the H-29/H-42/H-43 spread is as much an artifact of collapsing compatible facts into rivals as it is genuine uncertainty. Is the answer on the list? The most likely real-world truth is the conjunction H-42+H-29+H-43 (“eggs raise TMAO postprandially, not at fasting steady state, and this is probably benign in normal-renal people”), which no single member expresses; the residual H-56 is defined narrowly (a chronic steady-state elevation) and so does not catch that conjunction — a real carving gap, and an unlisted conjunction here is no more consequential than the listed poles, only better-fitting. COI cuts favourably: the fasting-null (Wilcox) comes from the Hazen/Tang lab that built the TMAO-harm hypothesis, so the eggs-null runs against author interest and is strengthened, not discounted. Miller (n=6) carries the postprandial-rise claim on very thin data.
What would help
- A chronic feeding study measuring both postprandial and fasting TMAO with outcome/surrogate markers in the same subjects — would collapse the H-29/H-42/H-43 split — exists, unread (Lemos/DiMarco partially corroborate the null).
- Egg/choline TMAO response in impaired-renal subjects (tests H-30, deliberately untested by Wilcox) — does not exist.
- A dose-response larger than Miller’s n=6 — exists, unread.
Confusions and contradictions
The two RCTs read opposite at face value — Miller: eggs raise TMAO; Wilcox: eggs do not — and the analysis resolves this through the postprandial/fasting distinction (O-105), which is probably right but means neither trial measured the quantity that bears on chronic harm (fasting steady state over months, with outcomes). The contradiction is reconciled, not eliminated, and the near-tie between H-29 and H-43 reflects that the data cannot say whether the operative fact is “no chronic exposure” or “chronic exposure but harmless” — a distinction that only matters at all if HC-8 makes TMAO causal.