Built on evidence, not marketing.
Every score, range, and recommendation in ByoMap is grounded in peer-reviewed research. Here's exactly how we turn your data into actionable intelligence.
7 Validated Clinical Scores
Tier 1 — Peer-reviewed, formula-based
These aren't proprietary black boxes. Each score uses published formulas from clinical research that physicians already rely on. We calculate them automatically from your biomarker data.
HOMA-IR
Insulin resistance index
Fasting Insulin × Fasting Glucose ÷ 405
< 1.0 optimal, > 2.5 resistant
Matthews DR et al., Diabetologia, 1985
TG/HDL Ratio
Atherogenic dyslipidemia marker
Triglycerides ÷ HDL Cholesterol
< 2.0 ideal, > 3.5 high risk
NCEP ATP III Guidelines, 2001
ApoB/ApoA1
Cardiovascular risk predictor
Apolipoprotein B ÷ Apolipoprotein A1
< 0.7 optimal, > 0.9 elevated
INTERHEART Study, Yusuf et al., Lancet, 2004
FIB-4 Index
Liver fibrosis staging
(Age × AST) ÷ (Platelets × √ALT)
< 1.30 low risk, > 2.67 advanced
Sterling RK et al., Hepatology, 2006
Omega-3 Index
Cardiovascular & brain health
EPA + DHA as % of RBC membranes
> 8% optimal, < 4% high risk
Harris WS, von Schacky C, Prev Med, 2004
Non-HDL Cholesterol
Total atherogenic lipid burden
Total Cholesterol − HDL Cholesterol
< 130 optimal, varies by risk
ESC/EAS Dyslipidemia Guidelines, 2019
Homocysteine
Methylation & cardiovascular marker
Direct measurement (µmol/L)
< 10 optimal, > 15 elevated
Refsum H et al., Annu Rev Medicine, 1998
Health Index
Tier 2 — Heuristic composite score
Your Health Index is a weighted composite of all four data modules. Unlike the clinical scores above, this is a heuristic — a useful signal, not a diagnosis. It uses severity-weighted percentile scoring, not flat penalties.
Biomarkers
200+ blood markers weighted by clinical severity and deviation from optimal
Gut Microbiome
Diversity, pathogen load, SCFA producers, and beneficial/harmful ratios
DNA
Genetic risk factors and predispositions from validated associations
Trends
Direction of change — improving markers boost your score, declining ones lower it
Biological Age
Biomarker-derived age estimation
Your biological age is estimated from 16 blood biomarkers that correlate with aging. Each marker contributes a weighted delta from age-adjusted optimal ranges.
We apply a 0.65× dampening factor to prevent extreme results, cap individual marker contributions at ±4 years, and total deviation at ±8 years. Markers with U-shaped optimal ranges (like SHBG) are scored for both high and low extremes.
This is not a clinical aging clock like Horvath's epigenetic clock — it's a biomarker-derived estimate. It's most useful as a relative tracker: are your markers trending younger or older over time?
Methodology
Ethnicity-Specific Ranges
Why one-size-fits-all is dangerous
Different ethnic populations develop disease at different biomarker thresholds. Standard lab ranges, built on narrow population averages, can miss early risk in many groups. The 2026 GenomeIndia study confirmed this at scale: 34% of Indian genetic variants are absent from global databases, and European polygenic risk scores lose up to 93% accuracy for Indians. ByoMap applies ethnicity-adjusted optimal ranges backed by population-specific studies for South Asian, East Asian, African, European, and Middle Eastern backgrounds.
Currently supported: South Asian, East Asian, African, European, Middle Eastern. Range resolver takes the tighter of demographic vs ethnicity-specific ranges. Sources audited against Tietz, NCEP ATP III, ADA 2024, KDIGO, ATA, ESC, WHO, ACG. GenomeIndia (2026) findings on LPA, LDLR, and APOB variant prevalence in Indian populations further validate tighter South Asian thresholds.
GenomeIndia: Why This Matters
9,768 whole genomes across 83 Indian populations (2026)
The GenomeIndia project — India's largest genomic study — proved what we've been building for: generic Western health tools fail Indian populations. ByoMap's ethnicity-adjusted ranges and personalized scoring are the direct answer to these findings.
34% of Indian genetic variants discovered by GenomeIndia are absent from gnomAD, 1000 Genomes, and GenomeAsia. Lab reference ranges built on those databases are structurally incomplete for Indian populations.
European-derived BMI polygenic risk scores drop from R²=0.097 to R²=0.007 in Indian populations — a near-complete collapse. Cardiometabolic risk algorithms must be recalibrated for South Asians.
The median Indian carries 4 pharmacogenomic variants affecting drug metabolism — for blood thinners, antidepressants, cancer drugs, and GLP-1 agonists like Ozempic.
17.5% of tribal Indians carry clinically pathogenic variants (vs 5.5% non-tribal) — including LDLR (familial hypercholesterolemia), BRCA2, and MYBPC3 (cardiomyopathy).
Source: Subramanian K, et al. “An Atlas of Indian Genetic Diversity.” medRxiv, March 2026. doi:10.64898/2026.03.20.26348801. GenomeIndia Consortium, 9,768 individuals, 129.93M high-confidence variants.
Range Audit
Key range improvements: LDL normalMax 130→160 (NCEP), toxic metals differentiated (lead optimal <3.5, normal <10), ALT/AST tightened per Prati 2002 criteria, homocysteine optimal <10 (Refsum 1998). Full audit automated via scripts/audit-ranges.ts --fix.
Transparency builds trust.
We show our work because your health decisions deserve a foundation, not a black box.
Try ByoMap Free