Deep Dive: Cannabis Genetics and Perceived Aroma

This 2,100-word Deep Dive explores, in detail, the effect of genetics on how humans perceive cannabis, particularly with respect to aroma and flavor. This article is a living document that is updated regularly based on the release of new research on this topic.

An October 2022 peer-reviewed research study entitled "Human Olfactory Discrimination of Genetic Variation within Cannabis Strains" published in the journal Frontiers in Psychology explores the issue of the cross influences between cannabis genetics and cultivars ("strains") and whether genetic inconsistencies in commercial samples "also display inconsistent aroma profiles."

The study's authors stated their goal as helping "the cannabis industry achieve better control of product consistency." "We sought to determine whether within-strain genetic variation in cannabis manifests as discrepant aroma production," reported the researchers.

The study reported that the number of "named cultivars is enormous: At least 600 have been described in the scientific literature" and that popular online databases list several thousand cannabis cultivars. This is critical for the burgeoning cannabis industry because "strain names are the basis of retail product identification in jurisdictions where cannabis is legal," reported the researchers.

The study noted that strain names are associated with "user-relevant attributes such as scent, flavor, appearance, and psychoactive effect" but that, ironically, they cannot be trademarked under U.S. law. It also reported another problem: Strain names "can be assigned capriciously [impulsively and unpredictably] by breeders, growers, and retailers." It explained that consumers hold expectations of particular strains and that this encompasses a desire for "consistent attributes," including aroma profile.

Consumers hold expectations of particular strains that encompasses a desire for "consistent attributes," including aroma profile.

"Ensuring consistency and quality in a psychoactive product is in the interests of consumers and the industry," reported the study's authors. They proposed that more knowledge of the "genetic and sensory variation between and within marketed strains" might allow patients and consumers to make "more informed purchasing decisions" while simultaneously permitting commercial cannabis cultivators to "achieve more consistent crops."

The researchers reported that recent studies identified significant genetic differences within individual cultivars, something that it called an "unexpected result given that commercial growers predominantly use clonal propagation [plant clones] to produce more uniform product." They noted that these genetic differences raise the possibility "that corresponding phenotypic variation—perhaps including measurable alterations in aroma—may be present as well."

The study reported that a more robust understanding of this genetic variation, both within and between popular commercial cultivars, is important if the emerging industry wishes to establish "quality attributes for commercial cannabis." The scientific investigation noted that it was designed to address the "physical basis of consistency" and that it focused on a single element of cannabis product quality: Aroma.

The study reported that "scent has a long history of use as a taxonomic marker in plants." It noted that Charles Darwin in 1876 "drew attention to correlations between floral scent and color" and how they might affect the "sensory abilities of insect pollinators."

The researchers noted that the cannabis industry is attempting to "broaden the quality focus to attributes other than THC content" and that this change in focus includes aroma. It reported that the link between scent and strain in cannabis "has become more salient" as the industry matures.

The research identified two major aroma profiles: 1) earthy, woody, and herbal group (Cluster A) and 2) citrus, lemon, sweet, and pungent (Cluster B).

The researchers cited a 2018 study that characterized "the aroma of dried flower across cannabis strains" and reported two major aroma profiles: 1) An earthy, woody, and herbal group that it labeled Cluster A and 2) a citrus, lemon, sweet, and pungent aroma profile that it called Cluster B.

The 2018 research studied samples of 11 strains, including two samples of a cultivar called Durban Poison, each purchased from a different licensed retailer. "Both samples of 'Durban Poison' were statistically grouped in Cluster B, even after new samples were re-examined. "These results suggest a certain degree of aroma consistency within named strains," reported the original study's authors.

The current study explained that a cannabis chemotype is the chemical profile of a strain and that aroma research to date has focused on terpenes, "the volatile compounds that are important in creating distinctive strain-specific aromas."

It described previous 2021 research that documented "variation in terpene profiles" between indica and sativa strains and a 2019 study that "proposed that the diversity of cannabis aromas observed today is a product of selection for scents believed to be associated with specific THC levels."

The study proposed that "trends in plant breeding may be driven in part by the role of aroma in consumer purchasing decisions" and that many consumers have particular expectations of "specific strain aroma profiles" with respect to psychoactive efficacy, including potency.

The study proposed that "trends in plant breeding may be driven in part by the role of aroma in consumer purchasing decisions."

"Given the economic importance of aroma and strain identity, we wondered if a genetically anomalous [varying] sample of a given strain would smell different from samples that were genetically cohesive [similar]," reported the study.

It employed "molecular genotyping and olfactory phenotyping techniques" to answer this question in a scientifically valid manner. It genotyped (determined the genetic nature) of 32 samples of cannabis from four strains. In addition, it evaluated 15 cannabis samples from the same four strains with "sensory judges who characterized the aroma using a standardized ballot." The scientists used the results of this evaluation to compare the aroma profiles of different genetic profiles within each strain.

The study's authors reported that they sought "to determine whether within-strain genetic variation in cannabis" manifests as different and discrete aroma profiles. "For this study, we gathered multiple samples of four cannabis strains from different sources in Colorado to be analyzed for genetic similarity."

The study noted that the availability of particular strains and their scent descriptions were obtained online from Weedmaps in 2022 and "cross-referenced with published data" to determine four candidate strains of cannabis. Durban Poison was selected as a representative of Cluster B (citrus, lemon, and pungent group) and OG Kush and Mob Boss were chosen as examples of Cluster A (earthy, woody, and herbal).

The researchers purchased 32 retail cannabis samples "from 21 state-licensed recreational dispensaries in six Colorado cities."

The researchers purchased 32 retail cannabis samples "from 21 state-licensed recreational dispensaries in six Colorado cities." The specific dispensaries chosen were based on strain availability. "A minimum of six samples [weighing two grams each] of each strain were collected," reported the study, from which DNA was extracted and analyzed.

The study reported results for these samples for genetic relatedness, sample clustering based on genetics, and a sensory analysis that involved human smelling (an "olfactory assessment") by study participants dubbed "odor judges."

"Odor judges rated each sample and the results were used to establish aroma profiles," reported the research. It noted that its objective with the olfactory assessment was to

determine if different aroma profiles existed among the same strains.

A total of 55 people comprised of 33 men and 22 women with a mean age of 30 participated in the study, all of whom were at least 21. "All but eight had purchased cannabis since recreational use was legalized in Colorado" in 2014. Ninety-one percent of study participants had smoked cannabis prior to participating in the study and 86 percent had purchased it via legal dispensaries.

The study's olfactory assessment found that the participants' perceptions of the samples clustered similarly to prior research. Its Cluster A featured predominant descriptors that included "earthy, herbal, and woody" and its Cluster B was described by study participants as "flowery, herbal, sweet, citrus, and earthy." The study's authors summarized that the "major olfactory divisions observed...are consistent with those reported previously."

The aroma evaluation revealed that the four samples of the strain OG Kush were located in Cluster A. In addition, "all three samples of Durban Poison were located within Cluster B," reported the scientists, who again stated that their results align with those of previous research studies on the topic.

"Blue Dream was found to range from pungent, chemical, and skunk-like to herbal and tea-like."

However, the scientists also discovered data that obscured this model, including one sample of "Mob Boss" that fell into Cluster A and another sample into Cluster B. They described their investigation of the aroma of the cultivar Blue Dream as the first of its kind, finding that two samples of this popular stain fell into Cluster A, while three were assigned to Cluster B.

"While all samples were characterized as herbal and earthy, those assigned to Cluster A were also highly rated for either sweet or flowery," noted the researchers. They found their OG Kush cannabis samples ranged in aroma from cheesy and woody to tea-like.

Blue Dream was found to range from pungent, chemical, and skunk-like to herbal and tea-like. As a result, two samples of this strain fell into Cluster A and three aligned to Cluster B. The Durban Poison cultivar samples yielded top-ranked descriptors of citrus, sweet, and lemon that were "consistent with the strain belonging to aroma Cluster B." The study noted that the Durban Poison samples also scored highly as herbal, earthy, and woody, " descriptors more typical of Cluster A.

The study explored the potential role of factors outside of genetics to help explain the "lack of consistent aroma profile cluster assignment" that it found in its analysis. "Two of the four strains examined raise questions about how non-genetic factors (e.g., differences in growth conditions, harvest time, post-harvest processing, etc.) can impact aroma."

"By focusing on the top-scoring odor descriptors within each strain, we observed substantial differences in aroma profile between the consensus samples and the genetic outlier," declared the study. It stated that this data suggests that genetic differences may produce different aroma profiles. "This suggests that variability in the neutral genetic markers we analyzed may be associated with phenotypic variations in aroma production," reported the research.

The study reported that "mono- and sesqui-terpenes are the most abundant volatile compounds in cannabis" and that they are responsible for the "characteristic odor of mature and dried [cannabis] flowers." It noted that differences in terpene composition have been noted among various cultivars of cannabis. "However, the association between specific terpenes and a strain's aroma profile remains speculative pending definitive studies using gas chromatography-olfactometry," reported the scientists.

"The association between specific terpenes and a strain's aroma profile remains speculative pending definitive studies using gas chromatography-olfactometry."

Interestingly, the researchers noted that a new family of chemical compounds called prenylated volatile sulfurs has been discovered in cannabis and that it may contribute the "skunky, diesel, or gassy" aromas that characterize many cultivars of the herb. "Volatile sulfur

compounds (VSCs) are present at low concentration, confirming the perfumer's dictum that chemical abundance is not a measure of odor impact: Less abundant molecules may make outsized contributions to aroma if they have low thresholds for odor perception," reported the study.

It explained that variation in terpene and VSC production, which directly impacts the aroma profile of a cannabis sample, "may have several non-genetic sources" that include cultivation conditions (such as "nutrient supply, light regimen, harvest timing, and post-harvest flower processing"). The research reported that something as seemingly incidental as the position of a flower along the stem may influence its aroma by changing the profile and ratios of terpenes and other aromatic molecules such as VSCs.

"As is the case for grapes and hops, phytochemical production in cannabis is likely influenced by a host of environmental factors," reported the research. It deemed that its data and analysis "lends further support to the role of non-genetic sources of

cannabis aroma."

The study concluded that its results confirmed "previous observations of within-strain genetic variability in samples purchased in the retail market." It found the results of its olfactory analysis to confirm the "high-level organization of cannabis olfactory space observed previously, namely that aroma profiles form two distinct clusters, one described as earthy, woody, and herbal (Cluster A) and the other as citrus, lemon, sweet, and pungent (Cluster B)."

The research also concluded that "the impact of non-genetic factors (e.g., differences in growth conditions, harvest time, post-harvest processing, etc.) on cannabis flower aroma remains under explored." It noted that variation in aroma between retail cannabis samples can be attributed to many factors, including differences in "olfactory discrimination and use of descriptors" by judges and differences in "terpene and cannabinoid content" resulting from harvest period, growth factors, and processing conditions.

Differences in aroma profile in cannabis retail products can be detected "by non-expert consumers and may impact their judgments of product quality and purchasing decisions."

The study concluded that it found "evidence that within-strain genetic variation in cannabis is associated with altered aroma perception" and that "phenotypic variation in odor production" warrants additional research and attention from the cannabis and hemp industries. The researchers concluded that differences in aroma profile in cannabis retail products can be detected "by non-expert consumers and may impact their judgments of product quality and purchasing decisions."

It stated that the development of aroma profiles "along with a metric for expected variation" could provide the emerging cannabis industry with a "useful standard for detecting departures from strain-specific aroma character" and ensuring product consistency. "Determining the relative impact of genetic and environmental factors on aroma production could help the cannabis industry achieve better control of product consistency," summarized the scientists.

View the original study.