TRAIN by Holi

Current Status

Our pilot phase is completed (4,000+ tracked workouts, proven VO2 max & body-comp gains) and $TRAIN token will fund the start of our BDNF trial in January 2026 (blood draws, lab analysis, and open-science infrastructure).

Summary

TRAIN by Holi is the first industrial-scale exercise-neuroscience engine producing multimodal datasets and licensable training protocols.

The first dataset is a head-to-head trial comparing velocity-focused vs. strength-focused training on chronic BDNF elevation and brain-health markers. Inside our fully owned Berlin GymLab (4,300+ tracked workouts, medical-grade equipment), 100 experienced members are randomized 1:1 into two 12-week arms. We perform 200 high-sensitivity plasma BDNF blood draws (baseline + week 12, standardized 24 h post-workout). Every participant receives continuous heart-rate monitoring during every session, and validated psychological batteries (e.g. PANAS, BRUMS, Cognitive Function Battery, Perceived Stress Scale, Pittsburgh Sleep Quality Index) at weeks 0, 4, 8, and 12. All raw data is published on-chain within 48 h. The superior protocol and full multimodal dataset become licensable IP governed and revenue-shared through the $TRAIN token.

Based on prior meta-analyses (n≈1,000+), we expect a between-group effect size in the range of g=0.35–0.55 for resting BDNF over 12 weeks. Our sample (n=100, 1:1 randomization) provides >80% power to detect group differences ≥0.4 at α=0.05. All plasma BDNF assays are processed by a certified clinical laboratory.

Problem

Today, no platform, app, clinic, or training system has comparative evidence for which training patterns reliably increase BDNF and brain-health markers over time. Velocity athletes and strength athletes have been arguing for decades about which style is better for the brain and longevity, yet no study has ever compared them directly in the same facility, with the same cohort, using gold-standard biomarkers. Existing BDNF research is almost entirely acute (<24 h), small-sample (n < 25), and ignores real-world variables like sleep, stress, and recovery — exactly what daily HRV, psychological batteries, and continuous heart-rate data solve. Trainers, apps, clinics, and longevity funds currently have zero comparative evidence to guide programming or investment.

Impact

TRAIN delivers the definitive answer: which training philosophy - explosive speed or maximal strength - drives larger, more sustained increases in resting BDNF, recovery capacity (HRV), mood, cognitive performance, and sleep architecture over 12 weeks. The multimodal dataset (BDNF + HRV time-series + heart-rate dynamics + psychometrics) instantly becomes the global benchmark for “brain-optimized training.” Because we own the gym and the data pipeline, we can run follow-up arms indefinitely (HIIT, Zone 2, concurrent, etc.) and keep compounding the most valuable longitudinal exercise-neuroscience dataset on Earth.

This study is designed for continuous iteration inside the DeSci standards. All assays, sensors, questionnaires, and workout streams follow DeSci’s reproducible-on-chain format (IP-NFT), enabling future cohorts (HIIT, Zone 2, concurrent, women-only, age-stratified) to reuse the same validated data pipeline without re-inventing operational overhead.