Personal Genomics Skill v4.4.0

Comprehensive local DNA analysis with 1600+ markers across 30 categories. Privacy-first genetic analysis for AI agents.

🆕 v4.4.0: 1000 GENOMES POPULATION COMPARISON & ANCIENT DNA

Transparent population comparison (not black-box percentages)
Ancient DNA signal detection (WHG, ANF, Yamnaya, Neanderthal)
Interactive dashboard with population frequency visualizations
Complete methodology documentation

⚠️ v4.3.0 focuses on ACCURACY AND HONESTY - improved uncertainty handling, PMIDs for all claims, and explicit limitations.

Quick Start

python comprehensive_analysis.py /path/to/dna_file.txt

⚠️ Important Limitations

Haplogroups are LOW CONFIDENCE - Consumer arrays cannot reliably call haplogroups. Recommend dedicated Y-DNA/mtDNA testing (FTDNA, YFull) for accuracy.
Ancestry shows ANCIENT SIGNALS, not modern ethnicity - Modern ethnicity percentages are unreliable. Instead we detect signals from well-characterized ancient populations (WHG, Neolithic Farmers, Steppe, Neanderthal, Denisovan).
PRS scores show RANGES, not point estimates - Polygenic risk scores have wide confidence intervals. Most conditions are 50-80% non-genetic.
Every marker has PMIDs - All claims are backed by literature citations linked to PubMed.

Triggers

Activate this skill when user mentions:

DNA analysis, genetic analysis, genome analysis
23andMe, AncestryDNA, MyHeritage results
Pharmacogenomics, drug-gene interactions
Medication interactions, drug safety
Genetic risk, disease risk, health risk
Carrier status, carrier testing
VCF file analysis
APOE, MTHFR, CYP2D6, BRCA, or other gene names
Polygenic risk scores
Haplogroups, maternal lineage, paternal lineage
Ancestry composition, ethnicity
Hereditary cancer, Lynch syndrome
Autoimmune genetics, HLA, celiac
Pain sensitivity, opioid response
Sleep optimization, chronotype, caffeine metabolism
Dietary genetics, lactose intolerance, celiac
Athletic genetics, sports performance
UV sensitivity, skin type, melanoma risk
Telomere length, longevity genetics

Supported Files

23andMe, AncestryDNA, MyHeritage, FTDNA
VCF files (whole genome/exome, .vcf or .vcf.gz)
Any tab-delimited rsid format

Output Location

~/dna-analysis/reports/

agent_summary.json - AI-optimized, priority-sorted
full_analysis.json - Complete data
report.txt - Human-readable
genetic_report.pdf - Professional PDF report
dashboard.html - Interactive visualization

New v4.3.0 Features (Accuracy Update)

Honest Haplogroup Reporting

LOW CONFIDENCE labels on all haplogroup calls
Explicit disclaimer that consumer arrays can't reliably call haplogroups
Recommendations for dedicated Y-DNA/mtDNA testing services
PMIDs for haplogroup marker sources

Ancient Ancestral Signals (Replaces Modern Ethnicity)

Western Hunter-Gatherers (WHG) - Mesolithic Europeans (~15,000-8,000 BP)
Early European Farmers (EEF) - Neolithic Anatolians (~10,000-5,000 BP)
Steppe Pastoralists - Yamnaya/Bronze Age (~5,000-4,000 BP)
Neanderthal Introgression - Archaic human (~50,000-40,000 BP)
Denisovan Introgression - Archaic human (high-altitude adaptation)
Shows "Signals Detected" not percentages
Includes time periods and trait contributions
Based on ancient DNA studies with PMIDs

PRS with Uncertainty Ranges

Percentile RANGES instead of point estimates
Confidence intervals based on marker coverage
Explicit interpretation guidance ("likely average", "uncertain", etc.)

PMIDs Throughout

Every marker has at least one literature citation
Clickable PubMed links in dashboard
Methodology & Limitations section in dashboard

Legacy v4.0-4.2 Features

Haplogroup Analysis (indicative only)

Mitochondrial DNA (mtDNA) - maternal lineage
Y-chromosome - paternal lineage (males only)
Migration history context
PhyloTree/ISOGG standards

Ancient Ancestry (scientifically grounded)

Detection of ancient population signals
Based on well-characterized ancient DNA
Includes archaic introgression (Neanderthal/Denisovan)

Hereditary Cancer Panel

BRCA1/BRCA2 comprehensive
Lynch syndrome (MLH1, MSH2, MSH6, PMS2)
Other genes (APC, TP53, CHEK2, PALB2, ATM)
ACMG-style classification

Autoimmune HLA

Celiac (DQ2/DQ8) - can rule out if negative
Type 1 Diabetes
Ankylosing spondylitis (HLA-B27)
Rheumatoid arthritis, lupus, MS

Pain Sensitivity

COMT Val158Met
OPRM1 opioid receptor
SCN9A pain signaling
TRPV1 capsaicin sensitivity
Migraine susceptibility

PDF Reports

Professional format
Physician-shareable
Executive summary
Detailed findings
Disclaimers included

New v4.1.0 Features

Medication Interaction Checker

from markers.medication_interactions import check_medication_interactions

result = check_medication_interactions(
    medications=["warfarin", "clopidogrel", "omeprazole"],
    genotypes=user_genotypes
)
# Returns critical/serious/moderate interactions with alternatives

Accepts brand or generic names
CPIC guidelines integrated
PubMed citations included
FDA warning flags

Sleep Optimization Profile

from markers.sleep_optimization import generate_sleep_profile

profile = generate_sleep_profile(genotypes)
# Returns ideal wake/sleep times, coffee cutoff, etc.

Chronotype (morning/evening preference)
Caffeine metabolism speed
Personalized timing recommendations

Dietary Interaction Matrix

from markers.dietary_interactions import analyze_dietary_interactions

diet = analyze_dietary_interactions(genotypes)
# Returns food-specific guidance

Caffeine, alcohol, saturated fat, lactose, gluten
APOE-specific diet recommendations
Bitter taste perception

Athletic Performance Profile

from markers.athletic_profile import calculate_athletic_profile

profile = calculate_athletic_profile(genotypes)
# Returns power/endurance type, recovery profile, injury risk

Sport suitability scoring
Training recommendations
Injury prevention guidance

UV Sensitivity Calculator

from markers.uv_sensitivity import generate_uv_sensitivity_report

uv = generate_uv_sensitivity_report(genotypes)
# Returns skin type, SPF recommendation, melanoma risk

Fitzpatrick skin type estimation
Vitamin D synthesis capacity
Melanoma risk factors

Natural Language Explanations

from markers.explanations import generate_plain_english_explanation

explanation = generate_plain_english_explanation(
    rsid="rs3892097", gene="CYP2D6", genotype="GA",
    trait="Drug metabolism", finding="Poor metabolizer carrier"
)

Plain-English summaries
Research variant flagging
PubMed links

Telomere & Longevity

from markers.advanced_genetics import estimate_telomere_length

telomere = estimate_telomere_length(genotypes)
# Returns relative estimate with appropriate caveats

personal-genomics