Are Cannabis Strain Names Accurate? What Genetics Research Suggests

A strain name is the first thing you read on a New York menu and the last thing anyone verifies. It carries a reputation, a price expectation, and an implied promise about how the evening will go. A run of genomic research, including a large mapping study published this month, suggests that promise rests on looser ground than most shoppers assume.^[1]

This is not an argument that names are meaningless. It is an argument that they are the weakest line on the label, and that the strongest lines are sitting right underneath them.

What the New Mapping Study Found

A peer-reviewed study published in September 2026 in BMC Plant Biology produced the most detailed picture yet of genetic variation across a large cannabis collection.^[1] Researchers analyzed 271 Cannabis sativa accessions drawn from 29 populations across 12 provinces in China, covering wild, landrace, and cultivated plants.

Using whole-genome resequencing against a reference genome, they identified 2,699,120 single nucleotide polymorphisms, the single-letter differences in DNA that geneticists use as markers, along with 42,201 insertion and deletion variants. The plants sorted into five major genetic groups, and those groups corresponded closely to geographic origin rather than to anything a consumer would recognize as a strain family.

The team then measured THC and CBD in 90 plants and ran genome-wide association tests. The clearest signal for CBD sat near the end of chromosome 1, at CBDAS, the gene encoding cannabidiolic acid synthase. THC proved harder. Researchers found 323 potentially associated markers across 82 genes, but said they could not confidently name a definitive candidate gene behind the strongest THC signal.

That asymmetry is worth sitting with. Even with nearly 2.7 million markers in hand, the genetics behind the single number most consumers shop by remains partly unresolved.

The two strains above make the point without any genetics at all. They sit within a percentage point of each other on THC, and they are categorized at opposite ends of the spectrum. If potency alone determined the experience, that pairing would be difficult to explain. Their terpene lines and their reported effects diverge where the potency numbers converge.

Why the Name Itself Is a Weak Signal

The naming problem predates the current research and is well documented. A 2021 analysis in PLOS ONE sequenced 23 cannabis strains marketed in Canada and found 137,858 markers, of which 344 were sufficient to sort the plants into five groups that aligned with chemotype, meaning CBD dominant, balanced, or one of three THC dominant clusters.^[2] The authors were direct about the cause of the confusion. Decades of extensive and unrecorded hybridization, they wrote, have left the vernacular naming convention unreliable for identifying or selecting strains.

Earlier work reached a similar place from the retail side. A 2020 survey in Cannabis and Cannabinoid Research examined chemical profiles and genetic markers across Nevada medical cannabis samples and reported that commercial chemovar names misrepresent the underlying chemical and genetic diversity.^[3] Samples sold under one name frequently tested outside the chemical range associated with it.

None of this implies bad faith by growers or retailers. There is no registry that verifies a cutting matches the name attached to it, so a name records commercial lineage and market history, not confirmed genetics. Research suggests it should be treated accordingly.

The Category Label Has the Same Problem

If the strain name is loose, the indica and sativa tag above it is looser. A 2021 study in Nature Plants genotyped more than 100 samples and reported that plants labeled indica and sativa were largely indistinct at the genome-wide level, with the labels instead tracking variation in a small number of terpenes governed by terpene synthase gene clusters.^[4]

That finding is the empirical core of a point we have made before in our breakdown of indica, sativa, and hybrid categories. The category is a useful conversational shorthand and a poor predictive instrument. The chemistry underneath it does more work.

What to Read Instead

The practical consequence is a reordering of attention. Everything below the name on a New York label is measured. Cannabinoid percentages and the dominant terpene list come from laboratory testing of that specific batch, which means they describe the jar in your hand rather than a category it has been filed under.

Line on the label	What research suggests it is worth
Strain name	Weakest signal. Unverified, and inconsistent across producers
Indica, sativa, or hybrid	Loose. Tracks some terpene variation, not genome-wide difference
THC percentage	Measured and real, but a poor standalone predictor of experience
Dominant terpene profile	Strongest routinely available signal of chemical character
Your own recorded response	The only evidence specific to you

The habit that follows is unglamorous and effective. Note the terpene profile of flower you enjoyed, note the profile of flower that did not land, and look for the overlap. Over a handful of purchases that record becomes more useful than any name, because it is the one dataset built from your own response. Our piece on the 227 terpenes researchers have now counted in cannabis covers why that line on the label is a compressed signal rather than a complete inventory.

What This Research Does Not Say

These are plant genetics and plant chemistry studies. They sequenced DNA and measured compounds. They did not test how anyone felt, and they make no clinical claims. Early findings show that the naming system carries less information than its confidence implies. They do not show that any particular profile will suit any particular person.

The mapping study has its own boundaries, and the authors name them. The collection was drawn entirely from China, so the populations are not a proxy for the North American commercial gene pool. The cannabinoid measurements covered 90 plants out of 271. Broader genomic work, including a 2025 Nature study assembling a cannabis pangenome around the cannabinoid synthase genes, is still filling in how domesticated lines relate to wild ones.^[5] This is a field in motion, not a closed question.

The Bottom Line

Strain names are a vocabulary, not a specification. Studies indicate they emerged from decades of informal breeding with no verification layer, and genomic analysis keeps finding that the names and the chemistry drift apart. The information you can actually rely on is printed below the name, and it was measured in a laboratory. Greenline Apothecary's strains list the full terpene profile alongside category and potency across The Original, All In One 5g Jars, and Functional Flower, so the chemistry is there to read before the name does any persuading.

References

1. Genome-wide variation analysis of Cannabis sativa germplasm and genome-wide association study of cannabinoid content. BMC Plant Biology. 2026. doi:10.1186/s12870-026-09909-5. Summarized in: Martinelli A. Researchers Map Nearly 2.7 Million Cannabis Genetic Variants, Identify Gene Strongly Linked to CBD Levels. The Marijuana Herald. September 8, 2026.
2. Chen J, et al. Classification of cannabis strains in the Canadian market with discriminant analysis of principal components using genome-wide single nucleotide polymorphisms. PLOS ONE. 2021;16(6):e0253387. doi:10.1371/journal.pone.0253387
3. Reimann-Philipp U, Speck M, Orser C, et al. Cannabis Chemovar Nomenclature Misrepresents Chemical and Genetic Diversity; Survey of Variations in Chemical Profiles and Genetic Markers in Nevada Medical Cannabis Samples. Cannabis and Cannabinoid Research. 2020;5(3):215-230. doi:10.1089/can.2018.0063
4. Watts S, McElroy M, Migicovsky Z, et al. Cannabis labelling is associated with genetic variation in terpene synthase genes. Nature Plants. 2021;7:1330-1334. doi:10.1038/s41477-021-01003-y
5. Domesticated cannabinoid synthases amid a wild mosaic cannabis pangenome. Nature. 2025. doi:10.1038/s41586-025-09065-0