trune™ is built on a scientific reality: people respond differently. Our methodology combines nutritionist-validated decision rules, safety-first dosing guardrails, and iterative within-person evaluation to learn what works for each individual.
Multiple large studies show substantial interindividual variability in metabolic responses—even to standardized meals—supporting methods that learn within-person rather than relying solely on population averages.
Key evidence
Personalized prediction of glycemic responses demonstrates that the same foods can produce very different post-meal responses across people.
Key evidence
Large-scale "precision nutrition" work (e.g., PREDICT 1) observed large inter-individual variability in postprandial metabolic responses and developed models to predict those responses.
Averages can hide what's true for you.
We operationalize personalization as a structured process: define evidence-backed options, apply safety constraints, choose a small set to test, measure response repeatedly, and update based on observed within-person effects.
Our methodology aligns with evidence-based dietetics principles: integrate best available research, professional expertise, and the individual's goals/constraints. Nutritionists define the decision rules and safety constraints; the platform executes them consistently and documents changes.
Nutritionist Review Gates
Evidence synthesis
Protocol template creation
Contraindication rules
Dosing bounds
Review gates
Versioned updates
Aligned with Academy of Nutrition and Dietetics evidence analysis process
Evidence Grade Badges
For nutrients with established reference values, we use Dietary Reference Intake concepts—including Tolerable Upper Intake Levels (ULs). ULs are defined as the highest daily intake likely to pose no risk of adverse effects for almost all individuals; risk increases above the UL.
Dose Window Visualization
Adjust dose position
Contraindication Filter Demo
Toggle conditions to see how the protocol library responds:
Botanical ingredients can have meaningful effects but may carry rare adverse events or interaction risks. Our methodology applies stricter eligibility rules, tighter dosing bounds, and more aggressive monitoring for botanicals than for basic nutrition support.
In high-performance sport, supplement use requires risk analysis because contamination and inadvertent ingestion of prohibited substances is a known risk. Our methodology is informed by established sports nutrition frameworks that weigh evidence, safety, and permitted use—and treat risk as a first-class constraint.
Risk & Evidence Matrix
N-of-1 designs treat the individual as the unit of learning—well-suited to nutrition where responses vary. Reviews of N-of-1 approaches in nutrition discuss how repeated within-person measurement can inform personalized intervention decisions and how series of N-of-1 studies can be aggregated.
N-of-1 Logic
Measure
Collect repeated within-person data across cycles
Compare
Compare periods (baseline vs intervention) within the same individual
Decide
Keep what improved outcomes; remove what didn't
Repeat
Next cycle refines based on accumulated evidence
Iteration Log Template
Changed
Removed ashwagandha from evening protocol; increased magnesium glycinate from 200mg to 300mg.
Why
Ashwagandha showed no measurable improvement in sleep onset over 2 cycles. Magnesium showed dose-dependent improvement in sleep quality ratings.
Watchlist
Monitor GI tolerance at 300mg. If sleep quality plateaus, consider adding wind-down protocol support.
Adaptive intervention methods inform how to refine components under real-world conditions.
EMA research emphasizes that user burden is a major driver of dropout and lower compliance in mobile health. Our methodology therefore uses ultra-short, targeted prompts designed to capture context that passive data cannot—avoiding redundant questions and long questionnaires.
Burden vs Adherence
94%
Estimated adherence rate
Shorter prompts → better adherence → more reliable data
Redundant questions we avoid
How did you sleep?
Did you work out today?
How many steps?
Rate your mood (1-5)
Already inferred from wearable/device data
High-leverage prompts we ask
Travel today?
Late caffeine?
Unusual stressor?
Allergy symptoms?
Only what changes the decision
Personalization is only credible when it is auditable. Each update is documented (what changed, why, what we're watching), and the methodology explicitly accounts for uncertainty, missingness, and variability in real-world behavior.
Transparency Card (Example)
Changed
Removed evening L-theanine; added magnesium glycinate 200mg
Why
No measurable improvement in calm ratings after 2 cycles; magnesium shows stronger signal for sleep onset
Watchlist
Monitor sleep onset latency and GI tolerance over next cycle
Known Limitations
trune™ is a wellness platform and is not intended to diagnose, treat, cure, or prevent disease.