Use Your 23andMe or AncestryDNA Raw Data for Pharmacogenomic Medication Insights

Consumer DNA tests from 23andMe, AncestryDNA, and similar services were designed primarily for ancestry and genealogy. But the genotyping arrays they use also capture variants in pharmacogenes — genes that directly influence how your body metabolizes medications.

This means your existing raw data can be analyzed for pharmacogenomic insights without any additional testing. For a broader overview of what you can do with your DNA file, see what to do with 23andMe raw data. Below, we cover how consumer-grade pharmacogenomic analysis works, what it can tell you, how it compares to clinical testing options, and where its limits are.

What Consumer Genotyping Arrays Capture

Consumer DNA tests use genotyping arrays — chips that probe specific known genetic positions (SNPs) across your genome. A typical array tests 600,000 to 700,000 positions. Among these are hundreds of positions in pharmacogenes: genes like CYP2C19, CYP2D6, CYP2C9, DPYD, TPMT, NUDT15, SLCO1B1, and VKORC1.

Not every pharmacogenomic variant is covered on every array. Coverage depends on the platform and the chip version. However, the variants that define the most common and clinically relevant star alleles — particularly for CYP2C19, CYP2C9, DPYD, and TPMT — are well-represented across major consumer platforms. For specifics on which platforms cover which variants, see our data sources page.

Genes That Influence Medication Response

The pharmacogenes most reliably analyzed from consumer raw data include the following. Each plays a distinct role in how your body processes specific categories of medications.

CYP2D6 — The Most Polymorphic Drug-Metabolizing Gene

CYP2D6 metabolizes approximately 25% of all prescribed drugs, including many antidepressants, opioids (codeine, tramadol), tamoxifen, and antipsychotics. With over 130 identified allelic variants, CYP2D6 is the most polymorphic pharmacogene. Your metabolizer phenotype — poor, intermediate, normal, or ultrarapid — determines whether standard drug doses produce expected, elevated, or reduced drug levels.

Consumer arrays cover many CYP2D6 star allele variants, though gene deletions (*5) and duplications cannot be detected from array data alone. See our CYP2D6 poor metabolizer symptoms guide for clinical context.

CYP2C19 — Antidepressants, Clopidogrel, and PPIs

CYP2C19 metabolizes clopidogrel (Plavix), SSRIs (escitalopram, sertraline, citalopram), proton pump inhibitors, and antifungals. Consumer arrays typically cover the *2, *3, and *17 defining variants, enabling reliable phenotype assignment for most individuals. See our CYP2C19 and SSRI metabolism guide for details on how this gene affects antidepressant dosing.

CYP2C9 — Warfarin and Anti-Inflammatory Drugs

CYP2C9 affects warfarin metabolism and certain anti-inflammatory drugs (NSAIDs). The *2 and *3 variants, which reduce enzyme activity, are well-covered by consumer arrays. CYP2C9 poor metabolizers may require lower warfarin doses to avoid bleeding risk — one of the earliest pharmacogenomic applications in clinical medicine.

SLCO1B1 — Statin Myopathy Risk

SLCO1B1 encodes a transporter protein that moves statins (particularly simvastatin) into liver cells. The key rs4149056 variant reduces transporter function, leading to higher statin plasma levels and increased risk of muscle-related side effects (myopathy). This variant is available on most consumer genotyping platforms.

Other Pharmacogenes: DPYD, TPMT, NUDT15, VKORC1

DPYD determines how you process fluoropyrimidine chemotherapy drugs (5-fluorouracil, capecitabine) — certain variants can cause severe toxicity. TPMT and NUDT15 affect thiopurine drug metabolism, used in autoimmune diseases and certain cancers. VKORC1 interacts with CYP2C9 to influence warfarin sensitivity. Key risk variants for all four genes are typically present on consumer arrays.

How the Process Works: From Raw Data to Medication Insights

Getting pharmacogenomic insights from your existing consumer DNA data involves four steps:

1. Step 1 — Download your raw DNA file. Export your raw data from 23andMe (Settings → 23andMe Data → Download Raw Data), AncestryDNA, or another supported service. This is a text file containing your genotype data at hundreds of thousands of positions. See our step-by-step upload guide for detailed instructions.
2. Step 2 — Upload the file. Upload your raw data to DecodeMyBio. The pipeline automatically detects your genotyping platform and identifies which pharmacogenomic variants are covered.
3. Step 3 — Analyze pharmacogenomic variants. Your genotypes at pharmacogenomic positions are extracted, mapped to star alleles using PharmVar-standardized definitions (e.g., CYP2C19*2 is defined by rs4244285), combined into diplotypes, and translated to metabolizer phenotypes using CPIC-standardized categories.
4. Step 4 — Interpret drug metabolism pathways. Your metabolizer phenotype for each gene is matched against CPIC and DPWG clinical guidelines to identify relevant drug-gene interactions, generating your Medication Safety Report.

For a detailed walkthrough of each step, see How Pharmacogenomic Testing Works. For implementation specifics, see our methodology page.

Pharmacogenomic Testing Cost: How Options Compare

Pharmacogenomic testing cost varies widely depending on the type of test, who orders it, and whether insurance covers it. Here is how the main options compare:

For a detailed side-by-side comparison, see our pharmacogenomics testing comparison and GeneSight cost, insurance & alternatives guide.

Clinical Guidelines: CPIC and DPWG

The clinical value of pharmacogenomic analysis depends on the guidelines used to interpret results. DecodeMyBio maps phenotypes to guidelines published by CPIC (Clinical Pharmacogenetics Implementation Consortium) and DPWG (Dutch Pharmacogenetics Working Group) — the two most widely recognized pharmacogenomics guideline bodies.

Only interactions with CPIC Level A (strong evidence) or Level B (moderate evidence) are included in reports. This means every drug-gene interaction flagged has been validated through clinical studies, meta-analyses, or replicated pharmacokinetic data. Level A interactions have the strongest evidence base: genetic information should be used to change prescribing for these drug-gene pairs.

How Accurate Is Genetic Testing for Psychiatric Medications?

The accuracy of pharmacogenomic testing depends on two distinct factors: genotyping accuracy (whether the variants are read correctly) and clinical validity (whether the results are clinically meaningful).

Genotyping Accuracy

Consumer genotyping arrays are highly accurate for the specific variants they test — typically >99% concordance rates for individual SNPs. A 2020 study in Clinical Pharmacology & Therapeutics confirmed high concordance between consumer array-derived pharmacogenomic results and clinical-grade testing for common CYP2C19 and CYP2D6 variants. The FDA has approved 23andMe to report on certain pharmacogenomic markers, acknowledging the analytical validity of consumer genotyping for these specific variants.

What Consumer Arrays Cannot Detect

Consumer arrays test a predefined set of variants. They cannot detect:

• Rare or novel variants not on the array — a "Normal Metabolizer" result reflects only the tested positions
• Structural variants — gene deletions (e.g., CYP2D6*5) and duplications, which are clinically significant for CYP2D6 phenotype assignment
• Platform-specific gaps — different genotyping services use different array versions with different variant coverage

Clinical Validity

The drug-gene interactions reported by DecodeMyBio are backed by CPIC Level A and Level B evidence — the same evidence base used by clinical pharmacogenomics laboratories. CPIC guidelines are developed through systematic literature review and have been adopted by major medical institutions including the Mayo Clinic and St. Jude Children's Research Hospital. PharmGKB provides the curated evidence that underlies these guidelines.

Consumer-grade pharmacogenomic analysis is informational — not a clinical diagnostic. Clinical decisions should be confirmed through clinical-grade testing when the stakes are high. For a comparison of consumer-grade analysis and clinician-ordered testing approaches, see our pharmacogenomics testing comparison.

Pharmacogenomics by Condition

Pharmacogenomic testing is most relevant for medications used in psychiatry, cardiology, pain management, and oncology. We have dedicated guides for each major condition:

• Genetic testing for antidepressants — How CYP2D6 and CYP2C19 affect SSRIs, SNRIs, and tricyclics
• Genetic testing for ADHD medications — Why CYP2D6 matters for atomoxetine and why stimulants lack PGx guidelines
• Genetic testing for anxiety medication — SSRIs, SNRIs, and benzodiazepines in the context of pharmacogenomics
• How accurate is genetic testing for psychiatric medications? — What 23andMe data can and cannot tell you about medication response

Frequently Asked Questions

How many pharmacogenes can be analyzed from consumer raw data?

DecodeMyBio analyzes variants across multiple pharmacogenes including CYP2C19, CYP2D6, CYP2C9, DPYD, TPMT, NUDT15, SLCO1B1, and VKORC1. The exact number of reportable genes depends on the variant coverage of your specific genotyping platform. See our data sources page for platform-specific details.

Is consumer-grade pharmacogenomics the same as clinical PGx testing?

No. Clinical pharmacogenomic testing is performed in certified laboratories, often using sequencing or targeted panels that cover a broader range of variants including structural changes. Consumer-grade analysis is based on genotyping array data, which tests a fixed set of positions. Consumer results are informational and should not be used as a substitute for clinical testing when critical prescribing decisions are involved.

What are CPIC guidelines?

CPIC (Clinical Pharmacogenetics Implementation Consortium) publishes peer-reviewed, evidence-based guidelines for drug-gene pairs. These guidelines assign evidence levels and provide specific therapeutic recommendations based on metabolizer phenotype. DecodeMyBio uses only CPIC Level A and Level B interactions in its reports.

Can raw DNA data detect gene deletions or duplications?

No. Genotyping arrays test individual SNP positions but cannot reliably detect structural variants like whole-gene deletions (e.g., CYP2D6*5) or gene duplications. These structural changes require specialized techniques such as copy number analysis. This is one of the key limitations of consumer-grade pharmacogenomic analysis.

How much does pharmacogenomic testing cost?

Cost varies widely. Clinical panels like GeneSight cost ~$330 without insurance (often partially covered). Genomind is $300–$400. Lab-ordered clinical testing ranges from $250 to $1,000+. Re-analyzing existing consumer DNA data through services like DecodeMyBio costs $29–$59 with no prescription needed. See our comparison table above for a full breakdown.

How accurate is genetic testing for psychiatric medications?

Consumer genotyping arrays achieve >99% concordance for the specific SNP variants they test. The clinical recommendations are backed by CPIC Level A and Level B evidence. However, consumer arrays cannot detect all variants (particularly structural changes in CYP2D6), and results should be discussed with your healthcare provider. See our detailed accuracy analysis.

Which consumer DNA tests work for pharmacogenomics?

23andMe, AncestryDNA, MyHeritage, and FamilyTreeDNA all use genotyping arrays that capture pharmacogenomic variants. Coverage varies by platform and chip version, but the major pharmacogenes — CYP2C19, CYP2D6, CYP2C9, DPYD, TPMT, and SLCO1B1 — are well-represented across all supported services.

How many drug-gene interactions can be detected from raw DNA data?

DecodeMyBio identifies 150+ drug-gene interactions across 13 pharmacogenes from consumer raw data, covering medications including antidepressants, opioids, antiplatelet agents, statins, anticoagulants, and proton pump inhibitors. The exact number of reportable interactions depends on your genotyping platform's variant coverage.

References

1. Clinical Pharmacogenetics Implementation Consortium (CPIC) Guidelines. cpicpgx.org
2. PharmGKB — Pharmacogenomics Knowledgebase. pharmgkb.org
3. PharmVar — Pharmacogene Variation Consortium. pharmvar.org
4. FDA Table of Pharmacogenomic Biomarkers in Drug Labeling. fda.gov