← Back to Results

πŸ“Š Statistical Validation & Methodology

Understanding the science and accuracy behind your results

🎯 Confidence Levels for Your Results

Continental Ancestry (99.7% confidence)

99

Your major continental ancestry assignments are extremely reliable.

Regional Ancestry (94.2% confidence)

94

Country-level ancestry assignments are highly accurate for recent ancestry.

Ancient DNA Matches (87.5% confidence)

88

Matches to ancient populations are reliable but become less precise with age.

Deep Ancestry Timeline (73.1% confidence)

73

Very ancient migrations (>10,000 years) have broader uncertainty ranges.

πŸ”¬ Our Methodology

🧬 SNP Analysis

847,293 genetic markers analyzed across all 22 autosomal chromosomes plus X and Y chromosomes.

Algorithm: Admixture v1.3.0 Reference: 1000 Genomes Project Phase 3 Coverage: 99.2% of target SNPs

πŸ“š Reference Populations

2,847 reference populations from modern and ancient DNA databases worldwide.

Modern: 1000G, HGDP, UK Biobank Ancient: Reich Lab, Mathieson 2018 Total samples: 47,892 individuals

🎲 Statistical Methods

Bayesian inference with Monte Carlo sampling for ancestry proportions.

Iterations: 50,000 MCMC chains Burn-in: 10,000 iterations Convergence: Gelman-Rubin < 1.01

⏰ Dating Methods

IBD segment analysis combined with mutation rate clocks.

IBD minimum: 7 cM segments Mutation rate: 1.25e-8 per bp per gen Generation time: 29 years average

πŸ“ˆ Accuracy Benchmarks

Analysis Type Accuracy Rate Sample Size Validation Method
Continental Ancestry 99.7% 15,247 individuals Known genealogy validation
Recent Ancestry (5 generations) 96.3% 8,924 individuals Family tree comparison
Haplogroup Assignment 99.1% 12,156 individuals Sanger sequencing validation
Ancient Population Matches 87.2% 2,341 ancient samples Archaeological context
Migration Dating 74.8% 1,847 dated events Carbon dating correlation

πŸ† How We Compare

847K

Hidden Lineage

SNPs analyzed

665K

AncestryDNA

SNPs analyzed

640K

23andMe

SNPs analyzed

570K

MyHeritage

SNPs analyzed

Key Advantages:

βœ… More Ancient DNA
We include 40% more ancient reference populations than competitors
βœ… Better Geographic Resolution
Sub-regional accuracy down to historical counties and provinces
βœ… Advanced Algorithms
Latest machine learning methods from 2024 genetic research
βœ… Deeper Time Depth
Reliable ancestry estimates back to 200,000 years ago

βš™οΈ Technical Specifications

GENOTYPING ARRAY: - Illumina Global Screening Array v3.0 - 847,293 autosomal SNPs + 26,341 X-chromosome + 12,743 Y-chromosome - 99.7% call rate threshold - Hardy-Weinberg equilibrium p > 1e-6 QUALITY CONTROL: - Relatedness filtering: PI_HAT < 0.125 - Population stratification: PC1-PC10 adjustment - Batch effect correction: ComBat algorithm - Missing data imputation: IMPUTE2 with 1000G reference ANCESTRY INFERENCE: - Algorithm: Supervised ADMIXTURE with cross-validation - K-fold validation: 10-fold cross-validation for optimal K - Convergence criteria: Ξ”LogLikelihood < 0.0001 - Bootstrap resampling: 1000 iterations for confidence intervals ANCIENT DNA MATCHING: - f3 statistics for population affinity - f4 statistics for admixture testing - D-statistics for introgression detection - Coalescent simulations for demographic modeling HAPLOGROUP CALLING: - Y-chromosome: ISOGG 2024 nomenclature - mtDNA: PhyloTree Build 17 - Minimum coverage: 10x for Y-SNPs, 20x for mtDNA - Manual review of all novel variants

⚠️ Important Limitations

Geographic Bias

Our reference panels are strongest for European and East Asian populations. African and Indigenous American ancestry may be less precisely resolved.

Recent Admixture

Ancestry estimates represent averages across your genome. Individual chromosomal segments may show different patterns due to recent mixing.

Ancient Timeline Uncertainty

Dates older than 10,000 years have large confidence intervals (Β±2,000-5,000 years) due to limited calibration points.

Not Medical Advice

These results are for ancestry and genealogical research only. Do not use for medical decisions without consulting healthcare professionals.