Indica & Sativa: should we “batch categorize” cannabis according to active ingredients?

2nd edition: The original version of this article was published in October 2018.

Indica and sativa as categorization for the different modes of action of certain cannabis varieties or strains were already considered obsolete in 2018. For example, on September 29, 2018, the German Hanfverband headlined in a news article “Indica” and “Sativa” – just a fake? “

But headlines like SATIVA ≠ INDICA don’t even come close to getting to the heart of the problem. Because not only the terms indica and sativa suggest false similarities for cannabis products, but also individual strain names of cannabis. In this article I would like to explain how meaningful current categorizations of cannabis products are based on their family trees and why the batch categorization according to active ingredients can be a possible solution to current categorization problems.

Cannabis strains often don’t do what they imply with their name

The starting point for this article is a study by the Californian test laboratory owner Jeffrey Faber. After analyzing 1000 samples of unspecified varieties for 42 different possible active ingredients and their concentrations and ratios, it was clear to him: An exemplary OG Kush in Dispensary A has completely different terpenes than an OG Kush from Dispensary B. The Berry Kush in Dispensary B on the other hand could be very similar to the OG Kush in Dispensary A. As a customer, you don’t get what you associate with OG Kush at first and you can’t be sure whether the respective effects depend on the strain name.

In order to get to the root of this thesis, the cannabis company Bedrocan collaborated with the Canadian Dalhousie University for a scientific study focusing on genetic variations of different cannabis strains. When examining individual, selected cannabis varieties of different origins, analogous to the inconsistent terpene profiles of the same varieties from the Faber study, once again no match between the genotypes of these varieties was found in this study, and thus no statement derived from this about effects and active ingredient profiles according to certain genetic variations to be hit.

At this point, an in-depth look at the Canadian study follows in order to understand what specifically was investigated and found. For the impatient, you can alternatively skip this section and continue from the Genotype & Phenotype section. [click me]

The study of the genetic variation of cannabis strains in focus

With the help of a main component analysis, the Canadian university has examined to what extent different cannabis varieties differ in their genotypes from the genotypes of two reference cannabis strains “Indica” and “Industrial hemp” and to what extent the results deviate from the declarations of breeders and growers . The researchers also checked whether varieties with sativa proportions declared by the manufacturer can also prove this through their genotype or whether the classification via the DNA may not allow any concrete conclusions.

In order to go through the whole thing seriously, I have to go a little deeper into the details of the experimental setup. The Canadian researchers examined 81 “Marijuana” samples (THC-containing flowers for consumption, land races and commercial strains) and 43 “Hemp” samples (industrial hemp from Europe and Asia).

In order to be able to measure quantifiable differences, the genotypes of the individual samples, i.e. the genetic material in the DNA, were decoded. In biological terms, figuratively speaking, the genetic makeup of a plant (and also of all other living beings) hangs in small pearls on two opposing chains. There are 4 different colored pearls and each pearl has a connection to a pearl on the opposite chain.

However, one pearl can only create a bridge to a pearl with ONE certain color. The pearls are the so-called bases in biological jargon, two of them in connection make up base pairs. There are the four bases adenine, thymine, cytosine and guanine. These four bases can only form the two possible base pairs adenine-thymine and cytosine-guanine. The bases lie on two strands of DNA that wind in the form of a helix and are each connected by hydrogen bonds between two bases. Three consecutive base pairs form an amino acid. The specific sequence of the amino acids determines which genes and thus also characteristics a plant possesses. In other words: The individual combinations of consecutive amino acids from the beginning to the end of a DNA strand determine the later possible manifestations of characteristics of a plant. Important: The DNA only offers different possibilities of characteristic expressions. The actual expression of physiological structural features is also extremely influenced by the environmental conditions during cultivation.

Einzelnukleotid-Polymorphismus DNA

SNP = single nucleotide polymorphism. At this point there is an anomaly to the reference genetics in that a base pair replaces another base and thus influences the possible manifestations of the feature of this section.[source]

When examining the 124 samples in the study, it was found that, when analyzing their amino acid combinations, more than 14,000 single nucleotide polymorphisms can occur on the DNA strands across all samples. This means that there can be at least 14,000 anomalies for approximately 400,000,000 base pairs of the genetic code of cannabis. In context, this means that at least 0.0035% of all base pairs of cannabis DNA can occur in different forms depending on the variety and thus ensure genetic variations. [1]

While the 400,000,000 equally combined base pairs of the cannabis DNA ensure that the hemp differs from humans, mountain gorillas, ferns or hops, the at least 14,000 different bases, which, depending on the sample and presumably variety, ensure that the hemp differs from humans, Replace another sequence, so that differences between the individual strains of cannabis arise. So leaf length, courage to purplish, terpenes, growth habit, …

Comparing the DNA of industrial hemp with high THC cannabis for recreational and medical uses

After the decrypted amino acid sequences were available for the various samples of THC-containing cannabis and industrial hemp, the anomalies discovered were packed into a graphic:


SNT genetic variations in hemp indica sativa
[source] Based on a cannabis plant from Pakistan, which provides the indica amino acid combination reference, it is shown how high the genetic variance of the individual samples is compared to the reference. This dimension is shown on the Y-axis. The X-axis shows how much a genetics differs from an industrial hemp reference. Component 1 (PC1) makes up 9.7% of the total variance, component 2 3.6%. Each component summarizes various anomalies in one dimension.
What is clear, which can also be seen in diagram C, is the genetic differentiation of industrial hemp from cultivated or THC-containing cannabis. Both groups show clear peculiarities in their amino acid “interferers” (investigated single nucleotide polymorphisms summarized under PC1). In diagram (a) this becomes clear through the clear distribution of the two point clouds, in diagram (c) through the slight mixing of green and purple colored samples (assignment to two reference groups and mixed forms).

“Sativa” significantly more genetically diverse than “Indica”

It is interesting that labeled indica cannabis strains in diagram (a) can almost always be found in the range of amino acid anomalies referenced for Indica, while labeled Sativa strains show all possible anomalies.

The Indica declaration seems to work very well in practice. In the case of sativa, according to the evaluations in the study, there’s not much information value linked to the Sativa label from producers and breeders.

In diagram (a), the attentive observer will notice a weakness of the study: the industrial hemp sample has more in common with the THC containing cannabis called marijuana by the study authors than with the industrial hemp plants, while the cannabis indica sample, which is the reference for marijuana Category, should be closer to industrial hemp than to THC-containing cannabis. The authors of the study assume that either their samples were mixed up or that their categorizing into marijuana and hemp was wrong. The sample for industrial hemp came from a German gene bank and the study authors did not have enough information about whether the sample for the active ingredients analysis was actually cultivated like industrial hemp should.

Regarding differences in way of cultivation, just think of the separation of the females from the males in intoxicating hemp (sinsemilla), as this plays a major role in the genetic differences between industrial hemp and THC containing cannabis, according to the authors of the study. Cultivation in different climatic zones also leads to differences in active ingredients analysis. Another possible explanation is that the two reference genetics happen to be exceptional outliers.

Contrary to previous studies, the study authors emphasize that there are not only abnormalities in the area of cannabinoid-influencing genes, but also in all other areas of the DNA, relating to other characteristics. However, for the sake of completeness, the study authors explicitly mention that “Hemp” produces more CBD and “Marijuana” produces more THC and that this is also genetically determined and can be read from the genotypes of the different plants. But as I said, this observation is only part of all the differences that can be identified in the characteristics of the various samples; on the whole, the characteristics of cannabis plants are about much more than just the characteristics of the active ingredients.

From diagram (c) it can furthermore be concluded that industrial hemp tends to draw from a larger gene pool than THC containing cannabis, since the manifestation of mixed inheritance (heterozygosity) is higher in industrial hemp.

A funny comparison of the study authors: industrial hemp and THC containing cannabis differ in their genetic variance about as much as Europeans from Asians. So you can easily imagine what small but subtle genetic variances we are thinking about when considering the genetic differences between different varieties and that different sub-varieties of the same botanical species over time adjust themselves genetically to their environmental conditions and thus also to their genotypes change over time and thus form genetic clusters depending on the location. Huh, long sentence!

Comparing Genetic Differences Between Indica and Sativa

However, the scientists were not satisfied with these first graphics and created more to examine the more indica-sativa-specific differences genotype wise. They set up two dimensions again, to each of which they assigned a group of anomalies in the order of the amino acids. Since only THC cannabis is compared and listed here, the number of single nucleotide polymorphisms analyzed falls from 14,000 to 9,700.

genetic variation indica sativa
[Quelle] It is easy to read from the individual diagrams, and this is also emphasized by the authors in the study, that sativa and indica do have differentiated gene pools, but that these have become more and more mixed up over time due to frequent crossing and breeding and keeping family trees secret. In diagram (a) you can see isolated point clouds from sativa-labeled varieties, point clouds from indica-labeled varieties and a large mixed pile – the latter is home to the mixed genetics mentioned, many producers and breeders call them hybrids.
These varieties are also displayed as hybrids in the point cloud, recognizable by the third dimension above the blue-orange color gradient, which stands for a hybrid declaration. The hybrids here appear gray-blue and gray-orange.

In Diagram (c) it is noticeable that the higher the percentage of sativa, the greater the genetic variance. This confirms the previous observations that indica declarations allow a more precise statement about the actual genes than sativas with a high stated sativa percentage.

Diagram 2 is intended to show once again that declared sativas often hardly or not at all share the genetic variance of an ideal sativa. Durban Poison, Jamaican Lambs Bread and other strains share almost all of the anomalies with a suspected 100% Indica from Afghanistan, but are at the same time a long way from the nearest designated sativa. Here, too, it is noticeable that declared sativas tend to have greater deviations than declared indicas.

The genotypes of designated THC indica and industrial hemp varieties are closer together than declared sativa varieties to sativas

The study’s authors conclude that, according to their results, cannabis indica is genetically closer to industrial hemp than to “sativa”. It is also concluded that the higher the proportion of sativa in a variety, the greater the genetic variances compared to industrial hemp.

The bottom line is, and this is supported by another study, on the one hand the current designation of industrial hemp with Cannabis Sativa L. is not tenable based on the analysis of the gene pool.

On the other hand, a call is made to come up with a new method of classification that is not based on pedigree. The authors cite the current seed trade as a point of criticism of the industry, which suggests the distribution of a variety, but only leads to deviations from the original variety through new combinations and in the past was largely responsible for the fact that the samples available were from different seed banks such as Greenhouse Seeds or private sources did not bring the expected or declared results in the genetic analysis. Only cuttings would bring the necessary genetic stability that is suggested.

At the end of the consideration of the study, I would like to compare the results obtained with a paragraph from the encyclopedia of cannabis breeding by Mike MOD, published by Nachtschattenverlag. There it is written on page 25 ff. that indicas often pass on the strongest genes when inherited, because due to the relatively harsh environmental conditions in their areas of origin they were forced to strongly express characteristics such as resin content (= higher reproductive capacity) or early maturity (resistance to pronounced seasonal climates) and to pass on these genes dominantly.

Indicas would tend to show fewer phenotypes or genetic variances over several generations in order to preserve them for the survival strategy in hard areas. According to the author of the non-fiction book, Sativas often spread into several phenotypes in the first subsequent generation; starting from the seed, the results of specific genetics behaves like playing the lottery.

Thus, the many sativa outliers of the study cited above can also be substantiated with these observations. Because there is no doubt that all the varieties (orange) in the study declared with high proportions of sativa show a higher genetic variance than the indica varieties marked blue.

/Excurs end.

Genotype & phenotype

But back to our Kush strains: How can it be that two samples of OG Kush from different origins contain completely different terpene profiles? A short digression into the rules of inheritance theory.

The genes of a daughter plant result from the gene pools of the mom and dad plants. The papa gives the pollen with which the flowers of the mother are fertilized. Now both provide the most diverse characteristics, whereby each characteristic can be different for mom or dad.

Let us use the leaf length characteristic to illustrate this: Papa has long leaves, Mama has short leaves. During the meiotic cell division, the gene pools of mom and dad are mixed, with either dad or mom prevailing for each individual trait according to Mendel’s rules. If the characteristic leaf length is inherited in its form long dominant and the form briefly recessive, it would result for the daughter, for example, that she has a 75 percent probability of expressing long leaves.

But why the whole theory? Similar to the leaf length characteristic, the expression of cannabinoids and terpenes is determined by the gene pool of the plant and thus also its parent plants. It can be the case that when a female is recombined by pollinating a female with the pollen of a male, a terpene, which was still present in significant quantities in the mother plant, no longer occurs in the newly combined daughter (from the seeds produced) or occurs in a lower concentration.

Of course, it can also occur in much larger concentrations or a completely new terpene is formed in the daughter plant, which both father and mother plants were missing. Genetics is like playing the lottery – you never really know what you’re going to get. Parents-to-be will probably be more excited about this inevitable fact than plant-loving THC lovers who just want to harvest fat buds quickly. The expression of a characteristic can only be predicted from one generation to the next with a maximum of 75 percent probability. If two inherited genes are equally valid, heteroygosity can also occur, which means that not one characteristic prevails, but a kind of cross-section of two characteristics. For example, red flowers from mom and white flowers from dad become pink – you know that maybe from roses.

Classification of active ingredient profiles of cannabis via family trees is no longer possible

Bedrocan, the cannabis company that was the patron of the study discussed above, supports these considerations with a press release about the study. They take up the result of the study that a effect-wise classification via the pedigree of cannabis plants is not really possible for the reasons discussed so far and that different phenotypes and therefore different terpene profiles are to be expected at harvest when growing from seed. In addition, they point out that far from the hardly volatile family trees, it is very likely that the phenotypic expression of a plant would be largely determined by the genotype of this (basic assumption in genetics). This thesis is supported by all the American breeders who cut cuttings from the respective mother plant for a specific variety and also achieve phenologically constant results under constant environmental conditions. That would then have to be proven again with a follow-up study. Excerpts from the press release:

The deconvolution of the indica-sativa lineage showed a strong relationship between the chemical and genetic profiles, suggesting that the different terpene content of the species is hereditary and important to the identity of these two groups [indica and sativa]. It is likely that strains are classified by their distinct flavors, rather than their pedigree – which has a direct impact on the genetics of that crop.

Importance of the terpene profile in cannabis in relation to the mode of action

The biomass of a cannabis flower consists of a large proportion of green plant material, a smaller proportion of THC and, in rare cases, CBD and an even smaller proportions of other cannabinoids and terpenes. Terpenes only make up a single-digit percentage of the total biomass of a flower, mostly in the per mil range, but they also work and taste in significantly lower concentrations than, for example, THC. Certain terpenes such as myrcene have a calming / sedating effect. Again, others promote concentration and / or creativity. Certain terpenes are pain relievers, while others are known to decrease or even reverse the rate at which specific tumors mutate.

What is considered to be fairly certain: How a strain works depends particularly on the terpenes present. Ultimately, they also modulate / moderate the amount in which the THC is absorbed and, for example, how it is metabolized.

The terpene profile of a plant thus determines its mode of action in the patient or consumer. And we also associate certain modes of action with the terms indica and sativa. For example, we think of an indica as a sedative and a sativa as a happy activator. Indicas are more physical and sativas are even more psychedelic. So the traditional associations.

But where exactly is the flaw in the Indica – Sativa system?

Strains that we assumed in the past to look like an indica have been recombined over and over again by numerous breeders over the past few decades. An OG Kush, for example, was always recombined in that mom and dad OG Kush were bred to another, preferably better OG Kush (back-breeded / back-crossed). It is safe to assume that it was not just clones / cuttings that were used to spread a strain like the world famous OG Kush. Because they are by no means so inconspicuous to smuggle in, as they take up more pack size and, unlike seeds, are quite sensitive. For this reason, especially in densely populated Europe, seeds have established themselves as the first choice for carriers of genetic information, which has led to particularly great genetic diversity and variation in this particular area of the world.

These purely subjective selections of a variety like OG Kush by numerous growers ultimately led to numerous phenotypes of this variety and, secondly, to numerous OG Kush genotypes in the form of seeds and cuttings. If all American growers wanted to grow their OG Kush, which is grown very often, from the same gene pool, they would have had to concentrate purely on cuttings from controlled and identical origins, seeds should not have played a role.

In a Netflix documentary on cannabis, plants that are solely multiplied cutted are referred to as sexually frustrated plants. Because males no longer play a role in cuttings-based breeding of an already stabilized variety. These processes are commonplace in regulated areas such as California or in the case of CBD in Switzerland, for example.

Another assumption that I am subordinating to at least some of the growers: When selecting our exemplary OG Kush, not every grower decided which OG Kush will continue to be used for the next generation and which have to kicked out based on the terpene profile. There are also characteristics such as yield or resistance etc. that breeders choose, which the patient ultimately does not notice.

Since terpenes have not been known for that long and laboratory tests are only selective available in luxury dispensaries in Las Vegas or on the Culture Boat in Utrecht, I definitely don’t take that offense at anyone. It just means that it takes a lot of effort to determine the terpene mix of each plant in the selection process in order to have a factual basis for decision-making for the following crosses. So all in all it is very understandable why an OG Kush in Dispensary A does not correspond to the OG Kush in Dispensary B and why Kush Berry from Dispensary B can taste and work like OG Kush from Dispensary A.

Indica and Sativa: we have ruined ourselves over time.

A thought, which has been in my head for ages: Why are so many seed banks so reluctant to inform that the seeds of one variety can sprout out of the earth in very different phenotypes and thus different terpene profiles? The study discussed has shown: Because in the past, selection was often not made according to the indica or sativa-determining terpenes, one cannot claim today that an indica actually has the same effect as an indica, because a seed always does it’s own subject to the genetics and thus trait expression lottery.

At this point, we can of course continue to pretend that the ambitious home grower is always getting a nice lemon-sativa phenotype from his Super Lemon Haze seeds. Or we look reality in the eyes and test our weed much more often for active ingredient levels and profiles. I would like that for recreational use too, but especially for medical applications. Because standardized THC values ​​alone are of no use if individual batches of, for example, Pedanios 18/1 differ in their terpene profiles. As of now, one batch of Pedanios 18/1 can help the patient very well, while the next one causes more problems than, for example, relieving pain. It has not yet been stipulated in any regulatory framework that different batches / batches of the same variety have to be the same in their terpenes within certain tolerances.

The solution: Test each batch for terpenes and categorize, select and optimize cannabis varieties for their specific terpene profiles.

That sounds like a lot at first. But terpenes are essential for various effects of cannabis. It is not for nothing that patients or recreational users have very different experiences with, for example, full-spectrum CBD oils or CBD monopreparations. This phenomenon is also known as the entourage effect. This effect describes the interactions between different cannabinoids and terpenes. Means that an isolated molecule can have a completely different effect alone than in combination with another.

Quality-conscious traders, whether in a legal or illegal, recreational or medical context, should have each batch tested for cannabinoids and especially terpenes by a laboratory.

Incidentally, batch means: the descendants of a single mother plant, planted at the same time under the same environmental conditions and fertilizer recipes.

Every genetics grown in a cultivation facility should be analyzed for terpene profiles at least once. Once you have determined the terpene profiles, you can already vaguely anticipate the effects of the batch tested. Varieties with more than 0.5% myrcene in the total biomass are, for example, under the more calming / physical-looking varieties of the spectrum.

With a mixture of many terpenes or a fairly even distribution of these, it is of course more difficult, because the individual terpenes form complex interactions. In this regard, more detailed research must be carried out into how the individual terpenes and in which combinations affect people or certain groups. In most cases, however, varieties contain “only” one to five dominant terpenes and using empirical values and a large collection of data, the determination of the effect can be continuously specified. We just have to start and keep going with measuring and cataloguing.

Cannabis batch categorization as a game changer?

In the United States, terpenes are now having a major impact on the entire cannabis industry. Terpenes now form the basis for selecting new varieties. After all, in California you don’t want to just get high / stoned, you want to know best beforehand whether you are going to get high or stoned. There are products there that are recommended for working or being creative, while others are supposed to induce a deep sleep. Everything based on the specific terpene profiles.

Of course, this whole categorization can either be based only on varieties that, according to the Californian clone-only philosophy, develop the same terpene profile over and over again, or on batch-based laboratory tests. Batch by batch must be tested to avoid a designated activity strain sending you to bed before 4:20 pm. This is the consequence of the development of the past few years, when in the context of illegality the selection was completely different and terpenes were not really known or the focus of the breeders.

Central body for the batch categorization of cannabis according to active ingredient content?

I imagine that all batches, at least for medical use, are sent to a central point. Each batch is then analyzed there and divided into categories based on the respective cannabinoid and terpene profile. Such a category, which the customer recognizes in the pharmacy, could be, for example:

  • stress relieving
  • appetizing
  • relaxing
  • promotes happiness
  • stimulates the appetite

Regardless of which “variety” such a categorization is based on – ultimately, modes of action cannot be derived from generalized genotypes, but solely by measuring the terpenes and cannabinoids contained. When a variety and a production facility are so advanced that they repeatedly produce the same terpenes and cannabinoid levels in each batch, the detour via a central point can of course be skipped (with appropriate random checks). There is still a long way to go until then, however, and I am curious to see if and when the terpene batch categorization will finally come.

Addition: Every clone-based variety (“clone only”) is also subject to genetic mutation

Even if a cannabis production company grows plants based on cuttings, that does not mean that the same terpenes are always formed in the same combination. The genetic basis of a cuttings is largely identical to that of the mother plant, but the genetics are only building instructions for the plant for all conceivable environmental situations. Depending on the environmental influences, the plant can react with different morphological reactions to certain influences with the help of this construction plan.

Therefore, a grow facility must ensure that the environmental influences always remain the same for a variety. But even if this is ensured, a variety based on cuttings is not immune to mutations and genetic changes. This is due to the molecular level. There will be an explicit article on the subject in the future. Broken down to the essentials, however, it can be said that the more cell divisions occur within a genetics, the greater the number of mutations. Cell division is the engine of plant growth and with each cell division the complete DNA of a cell is mirrored on the newly created cell. This process, known as mitosis, can, however, lead to errors such as linking bases with incorrect bases or single nucleotide polymorphisms. The phenomenon of non-disjunction, in which two non-sister chromatids are not separated during the metaphase of mitosis, can also lead to mutations.

Clone-only genetics also die over time

With increasing lifespan, genetics that are easily transferred to 100,000 cuttings in a commercial environment will develop more and more genetic variances from the original variety due to errors during DNA replication. Over time and more and more clones of a mother, but also with the age of the mother, which is actually only designed to last half a year, the number of mutations increases so much that the variety may be no longer being able to produce the usual standard.

Therefore, the terpene levels of cuttings-based cannabis strains should be tested again and again and, if the test values ​​differ, they should be categorized according to the above-mentioned system according to the active ingredient content or changes should be made to environmental parameters. The batch categorization would make it easier for Grow Facilities to raise cannabis without having to pay too much attention to the active ingredient consistencies. In this way, quality can be created without having to pay too much attention to standardized active ingredient contents, because more freedom is created between two successive batches. Just think of the differences between winter and summer grows.

Imagine what a batch categorization could mean for small growers too! With batch categorization, growers large and small could provide their share of the total cannabis demand. A dream for every social economist. A dream for the Spanish social club model or for friends of small businesses.