DNA Import

getbased can overlay genetic context on your blood work. Drop a raw DNA file from any major provider and getbased matches health-relevant SNPs to the biomarkers you're already tracking — so the AI can factor in your genetics when interpreting lab results.

Before You Start

You need a raw DNA data file from one of these providers:

Provider	File format	How to download
AncestryDNA	Tab-separated text	Settings → DNA → Download Raw DNA Data
23andMe	Tab-separated text	Settings → 23andMe Data → Download Raw Data
MyHeritage	CSV	DNA → Manage DNA kits → Download Raw Data
FTDNA (Family Tree DNA)	CSV	myFTDNA → Download Raw Data
Living DNA	Tab-separated text	Dashboard → Download Raw DNA

No AI provider is needed for DNA import — the file is parsed locally.

How to Import

1. Open the getbased dashboard
2. Drop your DNA file onto the page (or use the import button and select the file)
3. Review the import preview — SNPs are grouped by impact level
4. Click Import to save

The file is processed entirely in your browser using a Web Worker. Your DNA data is never transmitted anywhere.

What Gets Extracted

getbased doesn't store your entire genome. It scans your file for 42 curated SNPs across 10 categories and discards everything else.

Category	SNPs	Example genes	Related markers
Methylation	6	MTHFR, MTR, MTRR, CBS	Homocysteine, folate, B12
Iron	5	HFE, TMPRSS6, TF	Ferritin, transferrin, iron
Lipids	5	APOE, PCSK9, CETP	LDL, HDL, cholesterol, Lp(a)
Vitamin D	4	VDR, GC, CYP2R1	Vitamin D
Vitamin B12	4	FUT2, TCN2, CUBN	B12, homocysteine
Blood Sugar	4	TCF7L2, PPARG, ADIPOQ	Glucose, HbA1c, insulin
Sex Hormones	5	SHBG, CYP17A1, CYP19A1	Testosterone, estradiol, SHBG
Thyroid	3	DIO1, DIO2, TSHR	TSH, fT3, fT4
Bilirubin	2	UGT1A1	Total bilirubin
Fatty Acids	4	FADS1, FADS2, ELOVL2	Omega-3, EPA, DHA

APOE Haplotype

The two APOE SNPs (rs429358 + rs7412) are automatically resolved into the standard haplotype notation (ε2/ε3/ε4) rather than shown as raw genotypes.

How SNPs Are Selected

The curated list follows strict criteria — this is not a 23andMe-style trait report:

1. Must map to a tracked biomarker. Every SNP links to at least one marker in the getbased schema. Pure ancestry or trait SNPs are excluded
2. Known mechanism. The variant directly affects an enzyme, receptor, or transport protein in a biochemically understood way — not just a statistical GWAS association with a tiny odds ratio
3. Large effect size. Included SNPs have meaningful clinical impact (e.g., MTHFR C677T reduces enzyme activity to ~30%; HFE C282Y causes 10–20× iron overload risk in homozygotes)
4. Well-replicated. Findings confirmed across multiple databases (SNPedia, ClinVar, GWAS Catalog, PharmGKB)
5. Actionable. The genotype leads to a concrete recommendation — supplement methylfolate, monitor ferritin more closely, consider higher vitamin D targets, etc.

The core principle: only SNPs that help you interpret the blood work you're already tracking.

Effect Tiers

Each genotype is classified into one of three tiers:

Tier	Meaning	Example
🔴 Significant	Meaningful functional impact — worth factoring into your health decisions	MTHFR C677T AA (~30% enzyme activity)
🟡 Moderate	Mild effect — relevant when combined with other factors	MTRR A66G AG (modestly reduced B12 recycling)
🟢 None	Normal / wild-type — no impact	HFE C282Y GG (no hemochromatosis risk)

The import preview groups findings by tier so you can see significant results first.

Where Genetics Appears

Dashboard

A Genetics section shows your APOE haplotype and top findings grouped by category (methylation, iron, vitamin D, etc.), sorted by severity. The section is collapsible — click the header to expand or collapse. Initially shows up to 8 findings with a "show all" button for the rest. Appears after Key Trends charts. Each finding links to its primary study (PubMed) and a "more studies" page (SNPedia).

Detail Modal

When you open a biomarker's detail view, any relevant SNPs appear inline — showing the gene, your genotype, and what it means for that specific marker. SNPs are sorted by severity (significant first). Each includes links to the primary study and SNPedia.

AI Chat

The AI automatically receives your genetic context when interpreting lab results. SNPs with "none" effect are filtered out to reduce noise. Only SNPs relevant to the markers being discussed are included, unless you explicitly ask about genetics — then the full profile is sent.

Context Cards

SNPs route to relevant context cards (diet, supplements, sleep, etc.) so the AI can factor them into lifestyle recommendations.

Privacy

• Your DNA file is parsed entirely in your browser using a Web Worker — it is never uploaded or transmitted
• Only the 42 matched SNP genotypes are stored (not your full genome)
• Stored data includes: rsid, genotype, gene name, variant name, and effect classification
• DNA data is included in JSON exports and covered by encryption if enabled
• You can delete your genetic data at any time from the Genetics dashboard section
• A Genetics entry appears in the sidebar when data is present, for quick navigation

Re-importing

Importing a new DNA file replaces any existing genetic data for the current profile. You can re-import using the Re-import link at the bottom of the Genetics dashboard section, or by dropping a new file onto the page. This is useful if you switch DNA providers or want to update after downloading a newer raw data file.

Common Questions

What if my file isn't detected? getbased identifies files by their content headers, not just the filename. Make sure you're using the raw data export (a text or CSV file), not a PDF report or a health report download.

Why only 42 SNPs? Quality over quantity. Each SNP was selected because it has a well-understood mechanism, a large effect size, and directly helps interpret a biomarker you're tracking. Adding hundreds of low-confidence GWAS hits would create noise, not signal. Each SNP links to its primary research paper so you can verify the evidence yourself.

Can I add my own SNPs? Not directly. The SNP lookup table (data/snp-health.json) is curated and versioned — every entry must pass the five criteria above. Community contributions are welcome via pull request, but you must do the research first. Open an issue with:

1. The rsID and gene/variant name
2. Which biomarker(s) it maps to in the getbased schema (e.g., vitamins.folate, coagulation.homocysteine)
3. Mechanism — what enzyme/receptor/transporter does the variant affect, and how? (not just "associated with X in a GWAS")
4. Effect size — what is the functional impact? (e.g., "reduces enzyme activity to 30%" not "slightly associated with")
5. Replication — links to at least 2 independent studies or a meta-analysis confirming the effect on the specific biomarker
6. Actionability — what concrete recommendation follows from the genotype?

SNPs that fail any criterion will be declined. Common reasons for rejection:

• Synonymous variants with unconfirmed functional effects
• GWAS associations without a known biochemical mechanism
• Effects that only manifest on specialty panels (DUTCH, urinary metabolites) not standard blood work
• Weak or contradictory biomarker associations (e.g., studies finding opposite directions of effect)
• Variants whose pathway is already well-covered by existing SNPs (e.g., adding a third MTHFR variant when C677T and A1298C already capture the methylation story)
• Pharmacogenomic SNPs that only matter on a specific drug (e.g., methotrexate response)

Is this a medical diagnosis? No. Genetic information provides context for interpreting your lab results — it does not diagnose conditions. Always discuss significant findings with your healthcare provider.

mtDNA Haplogroup Import

In addition to autosomal SNPs, getbased can import mitochondrial DNA (mtDNA) mutation files to determine your maternal haplogroup and its mitochondrial coupling status.

Supported format

Some DNA labs (e.g., Living DNA) export mtDNA results as a simple CSV with one mutation per line:

263G
462T
1438G
10398G
13708A

Drop this file onto the app and getbased will:

1. Match your mutations against diagnostic markers from PhyloTree Build 17
2. Resolve your maternal haplogroup (e.g., J, H, U, K)
3. Classify your mitochondrial coupling status (coupled → uncoupled) using Doug Wallace's framework
4. Compare against your profile's latitude to detect environment-haplotype mismatch

Manual entry

If you already know your maternal haplogroup (from 23andMe results, FTDNA, or another service) but don't have a mutation file, you can enter it directly:

• Edit Client form — select your haplogroup from the dropdown under "mtDNA Haplogroup (maternal lineage)"
• Chat onboarding — in the DNA step, use the "Enter haplogroup" dropdown next to the upload button

Both paths auto-resolve the coupling classification and environment mismatch detection — no file needed.

Coupling classification

Based on Wallace 2015 (Cell), mtDNA haplogroups correlate with mitochondrial electron transport coupling efficiency:

• Coupled (L lineages) — efficient ATP production, low heat generation. Evolved in equatorial climates.
• Uncoupled (J, T) — more heat per calorie, less ATP. Evolved in cold northern climates.
• Intermediate — most other haplogroups fall between these extremes.

The AI uses this to contextualize your lab results, light exposure, cold tolerance, and environmental data.

Research framework

The coupling classification follows Wallace's mitochondrial paradigm — supported by cybrid studies and population data, but not universally accepted as clinical standard. Individual variation within haplogroups is significant.