Human Model · HM(t) at day 177

State-timeSynthetic demo data⟨ I, S_t, H_0:t, E_t, Θ_t, T, U_t ⟩ — belief, not fact; uncertainty shown, not hiddenSteering view

Localization — where is this individual biologically now?

Belief state b_t per dimension: value ± sd from within-person noise plus the observation model's semantic distance. Trajectory features (velocity, autocorrelation, recovery time) are state variables.

Dimension	b_t	dS/dt	autocorr	recovery	observation model
Therapy response	59 ± 9	-0.01/d	0.53	8 d	Imaging + tumour markers, interpolated · proxy · last read ≤ 14 d
Gastrointestinal function	25 ± 5.7	+0.43/d	0.54	7 d	Daily GI check-in · direct · today
Treatment toxicity	24 ± 8.5	+0.01/d	0.72	1 d	Labs, interpolated between draws · proxy · last draw ≤ 7 d
Inflammatory activity	31 ± 11.4	+0.42/d	0.41	7 d	CRP/ferritin + immune signature · inferred · weeks
Immune / metabolic	64 ± 11.5	-0.04/d	0.26	2 d	Metabolic panel + wearable · inferred · weeks
Nutritional sufficiency	62 ± 8.6	-0.16/d	-0.04	1 d	Weight, albumin, food log · proxy · days
Adherence	93 ± 4.5	-0.18/d	0.58	1 d	Dose + to-do completion · direct · today
Function / quality of life	61 ± 6	-0.18/d	0.42	2 d	PRO items · direct · today
Glycaemic control	125 ± 8	0/d	0	—	CGM daily mean · direct · today

High autocorrelation and slow recovery are early-warning signatures of a critical transition (Scheffer et al.), independent of the level.

Provenance — how did they arrive here?

The path H_0:t: baseline constraints, divergence, perturbations, actions and responses in order.

day 0baselineBaseline S17: MR phenotype; Intermediate metabolizer; butyrate 55
day 76perturbationPoor-sleep week (-3)
day 108perturbationDaily walks resumed (3.3)
day 109perturbationAntibiotic course (-4.7)
day 127divergenceTrajectory diverged from the population model
day 127noteTrajectory diverged from prediction
day 135perturbationAlcohol on dosing days (-1.8)

Now: between basins, heading towards failing.

Mechanism — which theories best explain the trajectory?

Versioned competing theories with supporting (E+), contradicting (E−) and missing-but-discriminative (E?) evidence. The contradiction engine lists what should NOT happen under each theory and whether it did.

T1 v1leading

confidence 68%

Mucosal barrier loss: drug-induced epithelial injury amplified by a FUT2-dependent glycan deficit; histamine and inflammatory load add on top.

E+

FUT2 non-secretor genotype (barrier glycan deficit)
DeepGene: Escherichia coli above range (LPS / zonulin)

E−

E? · would discriminate

Faecal calprotectin (would separate barrier from motility)
Serum DAO / histamine
Zonulin

Forbidden under this theory

✓ holdsGI should not improve while histamine intake stays high (GI velocity 0.32/day)

✓ holdsLow-residue meals should reduce GI within a week (not tested)

v1 baseline: genotype + phenotype

T3 v1competing

confidence 24%

Drug-exposure toxicity: pharmacokinetic over-exposure drives GI and systemic toxicity together; dietary and microbial factors are secondary.

E+

GI autocorrelation 0.71 (slow recovery)

E−

Toxicity within range

E? · would discriminate

Trough drug level
Dose-timing rechallenge (evening vs morning)

Forbidden under this theory

✓ holdsGI should not move with diet if exposure is the driver (0 dietary detractors with ≥ 80% confidence)

v1 baseline: pharmacogenomics

T2 v2falsified

confidence 4%

Bile-acid diarrhoea on a low-butyrate ecology: reduced SCFA production leaves bile acids unconjugated after fat-rich meals; GI burden tracks fat intake with a one-day lag.

E+

E−

Butyrate producers intact
No fat-load signal in the log

E? · would discriminate

Faecal bile acids
7α-hydroxy-4-cholesten-3-one (C4)
Rechallenge: 3 high-fat days under observation

Forbidden under this theory

✗ violatedGI should not rise on low-fat days (fat effect 0.5 over 43 low-fat days)

v1 baseline: microbiome → v2 dietary attribution

T4 v1competing

confidence 4%

Behavioural spiral: symptomatic days cut adherence and sleep, which raise GI and stress, which cut adherence further; the biology is a follower, not the driver.

E+

E−

Adherence intact
Sleep and stress stable

E? · would discriminate

Dose-taking timestamps vs symptom timestamps
Sleep-window rechallenge

Forbidden under this theory

✓ holdsGI should not worsen on days with full adherence and good sleep (insufficient paired days)

v1 from tracker trends

Forecast — which futures are likely under alternative actions?

30-day rollout of GI burden under the baseline model T and under each safe candidate action (effect ramps over onset). Dashed = prediction, never fact.

No action

GI 25 → 24

Intervention as measurement

Each past action as a measurement of this patient's transition dynamics: predicted ΔGI vs observed, prediction error ε_t, and the within-patient evidence patterns → individualized response confidence (observational, kept separate from randomized evidence).

No action started yet.

Decision — the Restore controller

Safe set computed first, then A* = argmax E[HealthGain + λ₁·InformationGain + λ₂·ConfidenceGain − λ₃·Risk − λ₄·Burden − λ₅·Cost]. λ₁ = 0.3 (a theory leads), λ₃ = 1 (stable enough). Execution stays with the clinician or the informed patient (EU MDR Class I).

Class	Action	Health	Info	Conf.	Risk	Burden	Cost	Utility
Observe	Continue passive collection, no perturbation A clean transition can be observed only without a new perturbation.	0	0.15	0.05	0	0.05	0	0.03
Measure	Faecal calprotectin (would separate barrier from motility) Discriminates T1 from its competitors (E?).	0	0.45	0.2	0.02	0.25	0.2	0.01
Maintain	Hold current regimen and content Protects nothing while the state deteriorates.	0.05	0.05	0.05	0.05	0.05	0	0
Test	Serum DAO / histamine Discriminates T1 from its competitors (E?).	0	0.45	0.2	0.02	0.25	0.5	-0.05

Excluded from the safe set

DT-01 Not indicated for the current state.
HY-01 Not indicated for the current state.
GI-04 Not indicated for the current state.
BR-03 Not indicated for the current state.
SC-12 Not indicated for the current state.
ME-01 Not indicated for the current state.
AD-05 Not indicated for the current state.
NP-02 Not indicated for the current state.
CR-01 Not indicated for the current state.

Evidence weights · w = Authority × Relevance × Persistence

Biological identity Intermediate metabolizer; FUT2, CYP2D6, GLP1R
Germline panel, baseline
0.86 (0.95·0.9·1)
High-authority clinical Toxicity 24, response 59
Labs / imaging, ≤ 14 d
0.77 (0.9·1·0.85)
Intervention response 0 within-patient responses recorded
ΔS after each action, accumulating
0.68 (0.75·1·0.9)
Deep biological state DeepGene retest day 90: damage 44/100, Shannon 2.76, butyrate 725 CPM, 4 opportunists above, 2 keystone absent
Synthetic DeepGene metagenomics + host DNA, sample 942724-R (compositional proxy)
0.33 (0.7·1·0.465)
Dynamic state GI 25, sleep 7.5 h, HRV 38, glucose 125
Companion + wearable, today
0.18 (0.6·1·0.3)
Exposure context Keep meals regular; three to four moderate portions spaced through the day; dosing per protocol; adherence 93%
Food log + dose log, event-level
0.1 (0.55·0.9·0.2)

LLIFE OS insight · the five questions, in prose

grounded · add an LLM key for prose

P-02694 is on day 177 of META-24 (MR, baseline S17): response declining, tolerance favorable, GI deteriorating. Over 14 days response moved -5 and GI +4; discontinuation risk is 24%. The state is between basins, heading towards failing, after diverging from the model on day 127.

Biology sets the gain: FUT2 non-secretor; butyrate capacity 55/100, diversity 3.7. Pushing it down: Escherichia coli 13.095% (LPS load) (-42.6), FUT2 non-secretor (-24.3), Bacteroides fragilis elevated (-16.8). Pulling it up: Protein (supports) (+18.7), Butyrate producers intact (+17.1). From the daily log: protein -3.5 GI next day.

No action has been tried yet.

No action is indicated now; the trajectory is stable relative to similar patients.