Genetic Profile

About Jon's Genome

I used to treat my mind like software: bugs to fix, features to ship. Genetics showed me some of the hardware constraints, but also where the knobs are.

Total Variants
AlphaMissense Scored
Critical + High Risk
Protective

The Quick Version

My methylation cycle runs at ~40% capacity. MTHFR C677T is homozygous, confirmed structurally pathogenic by AlphaMissense (score 0.6324). This single enzyme bottleneck cascades through DNA repair, neurotransmitter synthesis, and homocysteine metabolism. Riboflavin (B2) is the targeted support: the FAD cofactor that helps the impaired enzyme work better.

My nitric-oxide pathway leans low: I carry four NOS3 (nitric oxide synthase) variants, all homozygous — but they're common alleles, each only a moderate nudge toward less nitric oxide, not the rare combination I first assumed. Worth supporting anyway: L-citrulline and beetroot feed an alternative NO pathway.

My Phase II detox has one real gap: GSTM1 looks deleted (rs366631 tags a deletion carried by roughly half of Europeans), so I make one fewer glutathione-transferase enzyme — I'm still confirming the deletion directly from the raw sequence. The SOD2 variant I used to worry about turned out not to be a real risk for me. NAC and cruciferous vegetables support the backup pathways.

The good news: key protective systems are intact. I'm clear on the classic high-production TNF-alpha promoters (the −308 and −238 variants that show up in 30+ articles), carrying only one milder promoter. APOE ε3/ε3 (no Alzheimer's risk allele), FOXO3, Klotho, and DEPTOR (rs4871827, homozygous — an mTOR brake that also means fewer vascular complications from type 2 diabetes) longevity variants present, CRP trends low. My body converts plant nutrients poorly (BCO1, FADS1/2, VDR) — solved by choosing pre-formed supplements over precursors.

Beyond single genes, my whole-genome (polygenic) scores tell the most distinctive story: very high fluid intelligence (~99th percentile) and very high emotional sensitivity (neuroticism ~99.9th, confirmed by two independent scores), moderate ADHD, average autism (which hints it leans more on rare variants than common ones), and below-average anxiety — so my common-variant anxiety isn't elevated, and likely tracks more with ADHD than its own program.

The most striking thing in the whole profile is a gap between two scores that normally travel together. My fluid-intelligence score sits at the 99th percentile. My educational-attainment score sits at the 6th — and a second, independent education score agrees at the 11th. That's roughly ninety percentile points between two traits that are usually close relatives. The resolution is in what the education score actually measures: not intelligence, but years of schooling completed — which is intelligence plus a large non-cognitive remainder: persistence through tedium, delay of gratification, institutional compliance, tolerance for structure someone else designed. Read correctly, it says I have 99th-percentile cognitive machinery and bottom-decile scaffolding-and-compliance machinery. That's the genetic signature of an autodidact, and it quietly rewrites a story I carried for a long time about being a lifelong underachiever: the capacity was never the constraint. What was missing was non-cognitive scaffolding — and its absence is structural, not a matter of trying harder.

Two more scores make the neuroticism finding coherent rather than a one-score fluke: depressive symptoms ~91st percentile and subjective wellbeing ~36th — low wellbeing being the same signal read from the other end. Three independent scores pointing the same direction is an architecture, not noise. The boundary I drew above still holds, though: anxiety disorder specifically lands below median (~45th). The loading is for negative affect and rumination, not for an anxiety disorder.

The rest of the seventeen scores are a mix of confirmations and honest gaps. Openness ~73rd is the satisfying one: my measured Big Five openness came out at the 75th percentile, so a polygenic score and a self-report questionnaire built decades apart landed in the same place. Cognitive performance ~86th echoes the intelligence score from a different study. Height ~79th, cannabis use ~76th, and risk tolerance ~47th fill in the ordinary middle. And physical activity ~18th is the other loading-versus-expression gap on this page: a bottom-quintile predisposition to move, sitting against a training practice I actually keep. Genotype is a starting position, not a report card.

Does My Genome Match My Diagnosis?

In April 2026, I cross-referenced my 2,076,243 sequenced variants against the Psychiatric Genomics Consortium's largest GWAS studies for ADHD (225,534 participants) and Autism (46,350 participants), then computed validated Polygenic Risk Scores using LD-adjusted weights from the PGS Catalog. That AuDHD validation has since grown into a full polygenic profile spanning cognition, temperament, and lifestyle traits — each score shown below as a percentile against a gnomAD v4.1 European reference population.

Two findings in that wider profile ended up louder than the AuDHD scores I originally went looking for: the ~93-percentile-point gap between my intelligence and educational-attainment scores, and an internalizing cluster where three independent scores — neuroticism, depressive symptoms, and (inverted) subjective wellbeing — all agree. Both are unpacked in the summary above. The live profile below is the full set, and it's worth reading as a spread rather than a scoreboard: the interesting information is usually in where two related scores disagree.

Loading polygenic profile…

Explain It Like I'm Five

🔧
The Sticky Gearbox

My methylation cycle (MTHFR) runs at 40% efficiency, like a gearbox that needs extra grease. I add methyl-folate and riboflavin as the grease. Without it, things get stuck.

🔥
Warm Engine, Slow Drain

My dopamine engine runs a little hot and drains a little slow — I make more of it (TH) and clear it more slowly (COMT, DBH). Great for hyperfocus and idea-generation, rough for "just start the boring thing." It's part of the chemistry underlying my ADHD, and it's why external structure beats willpower.

🌡
The Sensitive Thermostat

My serotonin-and-stress thermostat reads the room a little hot — quicker to sound the alarm, slower to settle. But my whole-genome anxiety score is actually below average, so what feels like "anxiety" may be more about stress-reactivity and ADHD wiring. What helps most seems to be managing the load, not fighting the worry.

🤝
The Quiet Bonding Dial

My oxytocin "bonding dial" sits a notch low — empathy may run quieter and social situations can take more effort — but the same wiring comes with optimism and creativity. I bond through ideas and shared projects, not emotional radar.

🧘
Chill & Focus

My calm-and-focus chemistry runs a bit differently: slightly fewer CB1 "chill" receptors (CNR1, with a working-memory dip), slower breakdown of anandamide — the body's own "bliss" molecule (FAAH) — and glutamate-transport variants (SLC1A1) that nudge the anxiety and repetitive-thought dials. A niche corner, but it rhymes with everything above.

🌙
The Late Clock

About a dozen of my clock genes lean late — CLOCK and PER2 push me toward evening — and melatonin-receptor variants (MTNR1B) mean late meals hit my blood sugar harder, so eating dinner early is an intervention, not a preference. The twist: my whole-genome "morning person" score is ~93rd percentile, the opposite of those single genes. When a handful of famous genes argue with the sum of thousands, the polygenic sum usually carries the most weight — so the headline genes say "night owl" even as the genome-wide signal leans early. I hold it as genuine mixed evidence, not a settled call.

💧
The Clogged Pipes

My four nitric-oxide genes (NOS3) run a little low, so blood vessels relax less easily — but these are common variants with a moderate effect, not the rare combo I first assumed. Beetroot juice and L-citrulline give the vessels another way to relax.

🗑
The Missing Cleanup Crew

One detox gene (GSTM1) looks deleted — a common variant about half of Europeans carry — so I'm down one cleanup enzyme (I'm still confirming the deletion directly). NAC and cruciferous veggies back up the pathway.

💬
The Hidden B12

I'm a "non-secretor" (FUT2), so standard B12 blood tests can look normal even when I'm deficient. Sublingual B12 bypasses the absorption issue.

🍁
The Beta-Carotene Wall

Eating carrots doesn't give me vitamin A efficiently (BCO1). My body struggles to convert beta-carotene to retinol. I need pre-formed vitamin A from animal sources.

🚂
Big Engine, No Rails

Two of my whole-genome scores usually travel together, and mine sit about 93 percentile points apart: raw thinking power reads ~99th, while "years of school finished" reads ~6th. That second score isn't measuring how smart you are — it's measuring sit-still-and-finish: grinding through tedium and following structure someone else built. So the engine is enormous and the rails were never laid. It's why I taught myself nearly everything I'm good at, and why outside structure isn't a crutch for me — it's the track.

🎛
The Dimmer Switches

Single genes are on/off switches; traits like smarts, mood, and temperament come from thousands of tiny dials added up. Mine read very-high intelligence and very-high emotional sensitivity, moderate ADHD, average autism, and below-average anxiety. Seventeen of these dials are scored on this page, and the spread between them says more than any one reading does.

Featured Variants

These are curated by a simple rule: the systems that shape how I actually function — not the scariest-sounding variants. Most of what defines me is a moderate, polygenic tilt: no single dramatic gene, but several common variants nudging the same way. So the signal here is convergence, not severity. Predisposition ≠ destiny; it's a prior, not a verdict.

Actions & Experiments

Tier 1

Lifestyle Foundations

  • Sleep: 7-8 hours non-negotiable. MTHFR methylation and NOS3 vascular function both degrade with poor sleep.
  • Early dinner: Finish eating 3+ hours before bed. MTNR1B GG means late meals impair glucose metabolism more than average.
  • Beetroot / nitrate-rich foods: Dietary nitrate provides an alternative pathway for nitric oxide production, compensating for the 4 moderate NOS3 variants.
  • Exercise: Zone 2 cardio + resistance. Supports BDNF expression, nitric oxide production, and antioxidant enzyme activity.
  • Cruciferous vegetables: Broccoli, kale, Brussels sprouts provide sulforaphane which upregulates remaining GST enzymes, compensating for the confirmed GSTM1-null deletion.
Tier 2

Targeted Supplements

  • Riboflavin (B2): 100mg daily, the Swiss Army knife. FAD cofactor for MTHFR (AM confirmed), supports NOS3 via BH4 recycling, and is the FMO3 cofactor. One supplement, three pathways.
  • Pre-formed nutrients: Retinol (not beta-carotene) for BCO1, fish oil EPA/DHA (not flax ALA) for FADS1/2, vitamin D3 (not D2) for VDR/GC. The theme: bypass broken converters with the end product.
  • L-Citrulline + beetroot: 3-6g citrulline daily for 4 moderate NOS3 variants. Citrulline converts to arginine→NO via an alternative pathway. Beetroot provides dietary nitrate.
  • NAC / Glutathione: 600-1200mg NAC daily for Phase II detox gap. Precursor to glutathione, compensating for the confirmed GSTM1-null. Supports both antioxidant defense and detoxification.
  • Methylfolate + B12 + Creatine: Methylfolate 400-800mcg bypasses MTHFR. Sublingual methylcobalamin 1000mcg for FUT2/MTRR. Creatine 3-5g/day frees ~40% of methylation capacity.
Synthesis-ranked: These are the Top 5 meta-recommendations from the genetic synthesis. Prioritized by number of supporting variants and structural validation.
Tier 3

Discuss With Clinician

  • Homocysteine: Target <10 μmol/L. Functional marker for MTHFR methylation status. Test every 6-12 months. Elevated levels are an independent cardiovascular risk factor.
  • Blood pressure monitoring: Worth tracking given four (common, moderate) NOS3 variants. Regular home monitoring to validate L-citrulline/beetroot intervention. Target <120/80.
  • Omega-3 index: Target 8-12%. FADS1/2 variants mean plant omega-3 conversion is impaired. Validate that fish oil supplementation is reaching adequate EPA/DHA tissue levels.
  • 25(OH)D (vitamin D): Target 40-60 ng/mL. VDR/GC/CYP2R1 variants impair transport and receptor efficiency. May need higher D3 doses than population guidelines suggest.
  • MMA + B12: Use MMA (methylmalonic acid) as primary B12 marker because serum B12 is unreliable for FUT2 non-secretors. Sublingual methylcobalamin bypasses gut variability.
Important: This page is not medical advice. Genetics inform but don't determine. Consult professionals.

Methodology

Genetic data analyzed via live API against the Genetic Lifehacks variant database, enriched with AlphaMissense structural pathogenicity scores. Effect interpretations are based on published research, not predictive diagnostics.

Current Data Source Whole Genome Sequencing (Sequencing.com, 30x coverage)
Sequenced June 2024, imported March 2026 — 1,923 catalog variants scored from a 4.8M-variant whole-genome VCF
Prior Data Source 23andMe array-based genotyping (~640,000 SNPs)
Exported October 2023, superseded by WGS with quality scores
Reference Build GRCh38 (hg38 / Build 38)
Variant Database Genetic Lifehacks (1,925 unique variants, 2,927 article contexts across 19 topics), queried live via authenticated API
Structural Analysis AlphaMissense v2.0: 612 variants scored, 203 classified as likely_pathogenic
Deep learning model predicting missense variant pathogenicity from protein structure (Cheng et al., 2023, Science)
Analysis Date Variants scored with v2_3_palindromic — unresolved source-orientation guards, indel-aware scoring, and direction-aware composite tiers, AlphaMissense-enriched. Polygenic scores computed 25 May 2026.
Risk Classification Each variant's composite tier = its v2 risk level (how much the effect allele I actually carry matters, after a frequency cap that keeps common alleles from scoring high) × its AlphaMissense class, with a direction gate applied first (AlphaMissense measures how much a substitution disrupts the protein's structure — not whether the effect is good or bad, so a beneficial variant is never tiered Critical/High):
Protective = effect allele carried, but its annotated direction is beneficial (e.g. longevity, reduced disease risk)
Critical = high risk level + AM likely-pathogenic
High = high risk level (no / ambiguous AM), or moderate + AM likely-pathogenic
Moderate = moderate risk level (no / ambiguous AM), or high + AM likely-benign
Low = low risk level, or moderate + AM likely-benign
Uncertain = I carry it, but it can't be scored confidently (no population frequency to orient the allele).
= no personal verdict — either I don't carry the effect allele, or my genotype isn't available.
The Copies column shows how many copies of the effect allele my genotype carries (0/1/2), so "Not carried" and "Uncertain" are no longer both buried under one "Minimal" label. All of these stay off the headline counts.

Why a near-1.0 AlphaMissense score can still read "—" (not carried): an AM score describes one specific amino-acid substitution. If I carry the reference allele, I don't have that substitution — so there's no personal verdict no matter how alarming the score. That's exactly why several of my genes with near-1.0 AM scores (COL5A1, KIT) aren't featured: I carry the normal allele at all of them (Copies: Not carried).
Important Caveats Population frequencies vary by ancestry. Effect sizes are statistical averages, not individual predictions. Gene-gene and gene-environment interactions are not modeled. AlphaMissense scores predict structural impact, not clinical outcome. This is not diagnostic.

PRS Methodology: Polygenic scores are calibrated against gnomAD v4.1 Non-Finnish European allele frequencies. For the full calibration methodology, canonical percentile values, and regression QC, see the PRS Calibration QC Report.

⚠ Important Disclaimer

This page shares personal genetic data for educational and self-experimentation purposes. This is not medical advice. Genetic variants suggest predispositions, not diagnoses. Research evolves. Effect sizes vary. Individual responses differ from population averages.

Before making health decisions based on genetic information, consult with a qualified healthcare provider or genetic counselor who can interpret results in the context of your complete health picture.

"Genetics gave me a vocabulary for experiences I'd been having my whole life. Not excuses, explanations. And with explanations come options."

Understanding that my methylation runs slow doesn't mean I'm broken. It means I know which knobs to turn. Understanding that my stress response is sensitive doesn't mean I'm weak. It means I know why meditation isn't optional.


Predisposition ≠ destiny.
It's a prior, not a verdict. The code is written, but the configuration is mine.