Our methodology

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

Formula

Fasting Insulin × Fasting Glucose ÷ 405

Thresholds

< 1.0 optimal, > 2.5 resistant

Source

Matthews DR et al., Diabetologia, 1985

TG/HDL Ratio

Atherogenic dyslipidemia marker

Formula

Triglycerides ÷ HDL Cholesterol

Thresholds

< 2.0 ideal, > 3.5 high risk

Source

NCEP ATP III Guidelines, 2001

ApoB/ApoA1

Cardiovascular risk predictor

Formula

Apolipoprotein B ÷ Apolipoprotein A1

Thresholds

< 0.7 optimal, > 0.9 elevated

Source

INTERHEART Study, Yusuf et al., Lancet, 2004

FIB-4 Index

Liver fibrosis staging

Formula

(Age × AST) ÷ (Platelets × √ALT)

Thresholds

< 1.30 low risk, > 2.67 advanced

Source

Sterling RK et al., Hepatology, 2006

Omega-3 Index

Cardiovascular & brain health

Formula

EPA + DHA as % of RBC membranes

Thresholds

> 8% optimal, < 4% high risk

Source

Harris WS, von Schacky C, Prev Med, 2004

Non-HDL Cholesterol

Total atherogenic lipid burden

Formula

Total Cholesterol − HDL Cholesterol

Thresholds

< 130 optimal, varies by risk

Source

ESC/EAS Dyslipidemia Guidelines, 2019

Homocysteine

Methylation & cardiovascular marker

Formula

Direct measurement (µmol/L)

Thresholds

< 10 optimal, > 15 elevated

Source

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.

55%

Biomarkers

200+ blood markers weighted by clinical severity and deviation from optimal

20%

Gut Microbiome

Diversity, pathogen load, SCFA producers, and beneficial/harmful ratios

15%

DNA

Genetic risk factors and predispositions from validated associations

10%

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

16 gender-specific biomarkers (e.g. HbA1c, hsCRP, GGT, eGFR, Testosterone)
Age-adjusted optimal ranges per marker
Weighted coefficients per marker (0.3–1.5)
U-shaped scoring for hormones (SHBG)
Individual cap: ±4 years per marker
Dampening: raw total × 0.65
Total cap: ±8 years from chronological age
Noisy markers excluded (e.g. Basophils)

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.

Marker
Lab Normal
Ethnicity-Adjusted
Source
LDL Cholesterol
< 160 mg/dL
< 70–100 mg/dL
AHA/ACC 2018
ApoB
< 130 mg/dL
< 80–90 mg/dL
MASALA, INTERHEART
Non-HDL Cholesterol
< 160 mg/dL
< 100–130 mg/dL
ESC 2019
Triglycerides
< 150 mg/dL
< 100–120 mg/dL
Population-specific
HbA1c
< 5.7%
< 5.3–5.5%
ADA 2024
HOMA-IR
< 2.5
< 1.0–1.5
INTERHEART

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.

44M
novel variants

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.

93%
BMI PRS accuracy loss

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.

4
actionable drug-gene variants per person

The median Indian carries 4 pharmacogenomic variants affecting drug metabolism — for blood thinners, antidepressants, cancer drugs, and GLP-1 agonists like Ozempic.

17.5%
tribal pathogenic carrier rate

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

118
of 200+ markers audited against clinical literature
101
ranges updated from standard lab defaults
9
reference guidelines used (Tietz, NCEP, ADA, KDIGO, ATA, ESC, WHO, ACG, Prati)

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