At the heart of Canonic's advanced capabilities is computational genomics, which enables the innovative development of cannabis strains with a host of desirable features. Dr. Inbal Dangor, Head of Research and Development at Canonic, explains how this works and why patients play an important role in the process.
Take a deep breath: Canonic’s innovative cultivation process uses computational genomics and artificial intelligence to perform big data analyses in order to isolate and identify genetic markers that affect the phenotype of cannabis strains and chemical target traits. That’s it, we said it. And now, with the kind help of Canonic’s Head of Research and Development, Dr. Inbal Dangor, we’ll also understand what all this means.
“Computational genomics’ is a combination of two concepts”, explains Dangor, a Doctor of Plant Sciences from the Weizmann Institute. “Genomics is the science of the genome, which is the genetic load that every organism (living creature) has. The genetic sequence is written in a kind of language that has four ‘letters’ and different combinations of these letters create a kind of recipe according to which the body’s cells are active and function. There are, for example, genetic sequences according to which the cell produces a variety of proteins. These, in turn, play a variety of roles in flora and fauna. Depending on the type of protein produced, different traits in the organism will be expressed. So, in essence, genetic makeup affects the traits of every living thing.”
“Genomics delves into the genetic sequences and asks which of them confer such and such traits”, says Dangor. “For instance, what is the DNA sequence which produced a protein that is responsible for expressing a desired trait in cannabis? When it comes to a trait that is affected by only one letter in the genetic sequence, this is a relatively simple question. In practice, in most cases the situation is more complex because traits are affected by several different sequences so one has to look at the big picture. But how do you know where to look across a sequence of around a billion letters? This is an impossible task for a human being. This is where computational power comes into play.
Powerful computers tackle the challenge through statistical analysis. “These are many layers of information,” Dangor describes. “For instance, we add additional information to the genetic layer that we accumulated when we scanned many lines of cannabis and characterized their properties, for example, the concentrations of active ingredients present in them. It is possible to compare the sequences, cross with the data on the concentrations of the compounds and from this deduce the reasons for the differences between them.”
Ultimately, the goal of the process is to identify and isolate genetic markers. “Markers are sequences in the cannabis genome that we have identified that are related to the presence of a particular trait,” Dangor explains. “When they are present, we know we will find this trait in the plant. Namely, when carrying out a cultivation program in order to raise the level of a particular substance in cannabis, it is possible to know which plants contain the trait with the help of the genetic markers.”
The success rate goes up when you know what to look for
“Without markers, it’s more difficult to find what you’re looking for,” she adds. “For example, if I have a thousand seeds, I will have to grow them all and send them all for chemical analysis to check which of them has the desired profile. This is very inefficient. Therefore, our job is to identify and isolate as many genetic markers as possible. It is an ongoing process in which the information is constantly refined and new markers are added.”
Canonic inherited its knowledge and expertise from its parent company, Evogene, a world power in computational biology. “This gives us a huge advantage”, notes Dangor. “First of all, some of the computational tools come from there. Second, we have experience and knowledge accumulated about other crops and we leverage it. There are genetic sequences which are common to different plants so that this knowledge can also be applied to cannabis.”
“The more you know, the more reliable the analysis,” she continues. “As the system learns more sequences, it improves. The tools at the core of the system follow statistical rules according to which the more data there is, the lower the probability of error. Aside from the savings on resources and time, the success rate goes up when you know what to look for. We have seen this happen in other tumors that have undergone advanced cultivation as well.”
In this innovative process a place of respect for patients is reserved. “We are looking for feedback on an overall experience that is difficult to quantify”, says Dangor. “Therapeutic experience is influenced by lots of variables such as smell, appearance, taste and therapeutic effects. This is a very complicated and subjective matter. We therefore invite patients to join the Canonic G-mmunity, a community whose members will experience our products and share with us the therapeutic experience. We will cross-reference the information we receive from them with our database so that we can improve our development processes. And that’s the beauty of this process: despite all the technology and computing power, ultimately the patients themselves are the ones who will set the tone and contribute to the development of fine cannabis strains.