The Yamanashi Multi-omics Cohort (YMoC) enrolled 215 Japanese adults aged 30–70 with prediabetic glycemic profiles—fasting glucose 100–125 mg/dL and HbA1c below 6.5%—tracking them across three in-person visits over six months. Each participant underwent 75-g oral glucose tolerance testing with serial blood draws, liver elastography, anthropometry, and biospecimen collection spanning blood, urine, stool, and saliva for multi-omics profiling. Continuous digital phenotyping was captured via Fitbit Inspire 3 wearables and daily app-based questionnaires, while molecular data currently include genome-wide SNP array genotyping and longitudinal plasma proteomics in a subset.
This is a cohort design paper, not a findings paper—no biological discoveries are reported yet. Its value lies in infrastructure: most large biobanks sample participants infrequently and rarely integrate wearable digital phenotyping with dense omics layers simultaneously. YMoC's architecture, capturing HOMA-IR trajectories alongside proteomics and microbiome data in a tightly defined prediabetes window, positions it to identify early molecular signatures of diabetes progression before clinical thresholds are crossed. That window is arguably where preventive intervention has the greatest leverage. Limitations are significant: 215 participants is modest by biobank standards, the six-month follow-up window may be too short to observe meaningful metabolic transitions, and the Japanese-specific cohort limits global generalizability. As a preprint not yet peer-reviewed, the design claims remain unvalidated by independent scrutiny. Still, for precision prevention science, this represents a methodologically ambitious framework worth watching as omics data mature.