Cannabis & CBD Genetics from 23andMe or AncestryDNA Data (2026)
6 min read · Last reviewed: August 2026 · Vytautas Jazbutis
If you have ever wondered why cannabis affects you differently than the people around you — stronger highs, longer-lasting effects, unexpected side effects from CBD — the answer may be in the DNA data you have already uploaded. Your 23andMe or AncestryDNA raw file contains the exact genetic variants that determine how your body metabolizes THC and CBD.
On this page+
What Your DNA Data Already Contains
Consumer DNA chips from 23andMe, AncestryDNA, MyHeritage, and FamilyTreeDNA test hundreds of thousands of genetic positions. Among them are the key variants in four cannabis-relevant enzymes:
- CYP2C9 — the primary enzyme forming 11-OH-THC, THC's main active metabolite. In one study, people with two copies of the reduced-function *3 variant had roughly 3x higher THC exposure.
- CYP2C19 — a primary CBD-metabolizing enzyme alongside CYP3A4, and the main route to the active metabolite 7-OH-CBD.
- CYP3A5 — examined alongside CYP2C19 in a single small study of CBD-related diarrhea. The two genes could not be separated, and the finding is not clinically actionable (see below).
- CYP1A2 — induced by smoked (not vaped) cannabis through combustion byproducts. Affects caffeine, clozapine, and other medication levels.
Why THC Affects You Differently
The best-evidenced genetic factor is CYP2C9. In 43 healthy volunteers given oral THC, people with two copies of the reduced-function *3 variant had a median total THC exposure roughly 3 times higher than those with two normal-function copies, and showed a trend toward greater sedation. The *2 variant made no difference. Higher exposure may mean:
- The same dose produces a stronger, longer-lasting high
- THC stays in your system longer (relevant for drug testing)
- You may be more prone to negative effects like anxiety or paranoia
Reduced-function CYP2C9 variants are common in people of European descent. In a meta-analysis of population-scale sequencing data, the normal-function *1 allele accounted for about 82% of CYP2C9 alleles in Europeans, with *2 at about 12% and *3 at about 6% — so a substantial minority of people carry at least one reduced-function copy. (An earlier version of this article put that figure at 15–20%, which understated it.) These variants determine your metabolizer status — the same framework used for prescription drug metabolism.
Can Genetics Explain CBD Side Effects?
Not reliably, on current evidence. A 2026 secondary analysis of 33 healthy volunteers taking prescription-strength CBD (5 mg/kg twice daily) found diarrhea in 7 of 18 participants who were both CYP3A5 non-expressers and CYP2C19 intermediate/normal metabolizers, compared with 1 of 15 others (p = 0.0463).
That result is worth knowing, but it does not support a genetic test. The comparison group had a single case of diarrhea. CYP2C19 intermediate and normal metabolizers accounted for 7 of the 8 diarrhea cases, and the authors state the CYP3A5 contribution could not be analyzed separately because the groups overlapped — so this is not established as a two-gene interaction. They explicitly do not claim a cause-and-effect relationship, describe the pharmacokinetic differences as mild, and conclude that pharmacogenomic testing may hold questionable utility for predicting these adverse events. No study has tested this at the far lower doses in over-the-counter CBD products.
We go through this evidence in detail in CBD side effects and genetics.
How to Get Your Cannabis Genetics Report
DecodeMyBio's Cannabis & CBD analysis covers all four genes from your existing raw data — no new sample needed. It covers THC metabolism, CBD processing, your CYP3A5 and CYP2C19 genotypes with the research and its limitations, and smoked cannabis drug interactions. It reports genotypes and published evidence; it does not predict whether CBD will give you side effects. It is free to start ($59 one-time) and produces results in minutes.
You can also preview a sample Cannabis & CBD result to see the format before you start.
What the Report Does Not Cover
These results do not predict subjective experience (euphoria vs. anxiety), cannabis dependence risk, or tolerance effects. It does not recommend strains, products, or dosing. It is educational only and does not endorse cannabis use. Cannabis laws vary by jurisdiction.
References
- Sachse-Seeboth C, et al. Interindividual variation in the pharmacokinetics of Δ-9-tetrahydrocannabinol as related to genetic polymorphisms in CYP2C9. Clin Pharmacol Ther. 2009;85(3):273-276. PMID: 19005461.
- Dronabinol Therapy and CYP2C9 Genotype. Medical Genetics Summaries. NCBI Bookshelf, NBK564166.
- Zhou Y, Ingelman-Sundberg M, Lauschke VM. Worldwide Distribution of Cytochrome P450 Alleles: A Meta-analysis of Population-scale Sequencing Projects. Clin Pharmacol Ther. 2017;102(4):688-700. PMID: 28378927.
- US Food and Drug Administration. EPIDIOLEX (cannabidiol) oral solution — Prescribing Information (Section 12.3: metabolized in the liver and gut primarily by CYP2C19 and CYP3A4). 2018.
- Beers JL, Fu D, Jackson KD. Cytochrome P450-Catalyzed Metabolism of Cannabidiol to the Active Metabolite 7-Hydroxy-Cannabidiol. Drug Metab Dispos. 2021;49(10):882-891. PMID: 34330718.
- Etkins J, So GC, Lu JBL, et al. Genotype-Specific Safety and Pharmacokinetics of Cannabidiol in Healthy Volunteers. Clin Transl Sci. 2026;19(1):e70455. PMID: 41451876.
- Correction to “Genotype-Specific Safety and Pharmacokinetics of Cannabidiol in Healthy Volunteers”. Clin Transl Sci. 2026;19(2):e70506. PMID: 41685796.
- Anderson GD, Chan LN. Pharmacokinetic drug interactions with tobacco, cannabinoids and smoking cessation products. Clin Pharmacokinet. 2016;55(11):1353-1368. PMID: 27106177.
- Stout SM, Cimino NM. Exogenous cannabinoids as substrates, inhibitors, and inducers of human drug metabolizing enzymes: a systematic review. Drug Metab Rev. 2014;46(1):86-95. PMID: 24160757.
The CYP2C9 findings here come from a single study of 43 healthy volunteers, and the CBD side-effect findings from a secondary analysis of 33 volunteers on prescription-strength doses whose authors concluded that pharmacogenomic testing may hold questionable utility for predicting such events. These describe average differences between genotype groups, not predictions for any individual. This article does not recommend cannabis use or any dose.
DecodeMyBio provides informational pharmacogenomic and genomic insights only. This is not medical or nutritional advice. Always consult your healthcare provider before making medication or supplement changes.
How this content is created and kept current: Methodology · Editorial policy · Limitations