Featured News

Nick Jikomes, Principal Research Scientist At Leafly, Discusses Visual Profiling For Strains

Leafly is looking towards a more unique approach to cannabis identification

In today’s technology driven cannabis industry, data reigns supreme. While capturing that data is half of the battle, understanding that data’s importance and how to best utilize it is the real challenge. Enter Nick Jikomes. With a PhD in neuroscience from Harvard University, and a B.S. in genetics from the University of Wisconsin-Madison, Nick knows data. As Leafly’s principal research scientist, he has taken that data and helped to change the narrative on cannabis from 3 broad strain categories, Indica, Sativa, and Hybrid, to a visual exploration of a particular strains dominant cannabinoid and terpene profile.

We sat down with Nick to talk to him about understanding strains on a deeper analytical level in this exclusive interview.

MRR:                                 First and foremost thank you so much for taking the time to talk to me today. Now having a degree in neuroscience from Harvard is no small feat. With the educational background that you have, what was your initial draw towards the industry?

Nick Jikomes:                   I really like cannabis. *laughs* But it’s not just that. It’s like, you look at this stuff and it looks different, it smells different, it feels different, and so I always wanted to know what are the actual differences in cannabis strains and how can we create a reliable way to find exactly what you want. In an unregulated market, you just get whatever was available. So I had many experiences where I would find something that I loved but I had no idea what it was. Then one day it just would be gone and I had no way of like finding it again.

But even just as a scientist it’s super interesting, right? It’s a very chemically complex, diverse, and nuanced plant, so it’s super interesting to think about it not from just a product development standpoint in terms of creating products that people are gonna buy in a recreational store, but to also think about it from a product development standpoint on the medical side. I’m very interested in plant medicine and precision medicine and I think we’re moving, slowly but surely, on the medical and healthcare side to personalize medicine. That means medicines that are made specifically for you. Everyone’s got different genetics, everyone’s got a different biological setup. The flip side of that is the standard model for pharmaceutical development which is one drug in a pure form, with very specific pharmacology, and there’s limitations to that.

Essentially you can think about cannabis as sort of like a natural chemical factory. But it’s not just making one compound in a pure form, it’s making combinations of compounds that can act together in a complimentary fashion in order to achieve certain outcomes for the plant. So if you think about it from a medical standpoint, the best way to treat certain complex illnesses might not be one compound in a pure form. Rather, it might actually be a cocktail of compounds at precise ratios. It’s just very interesting that cannabis does amazing stuff to people both in terms of its effects on consciousness as well as just from a data science perspective. It’s fascinating to understand this and to be on the cutting edge of something.

MRR:                                 What was the process like conceptualizing how best to gather the data, and then how to get it into an easy to understand visual format?

Nick Jikomes:                   It was difficult. *laughs* Well let me be clear, it was relatively easy to understand how all of these strains are organized from a chemical perspective because you can do fairly standard data science stuff, and it’s a fairly standard analysis to understand that side of it. The challenge was how do you turn that into something that the consumer can use? So how do you create a way to visualize and represent the strains that’s both faithful to the data and the chemistry, but also usable to the everyday person. That was the challenge.

So we measure, you know, 25 or so compounds in these plants, cannabinoids and the terpenes. The algorithm uses a visual graph and puts two points together if they’re very close overall. So if two flower samples have a close overall profile, they are close together the graph and if they’re different, they’re far apart. So this is just pretty standard data science stuff. And if you just look at the data you’re like, okay, well most of the data’s over here, and there’s some stuff over here, and there’s more stuff over there, it’s separated. Then you just start to map out how it’s actually organized with respect to the chemical compounds that are inside of the stuff and the traditional way that the industry organizes things, right? Sativa, Indica, Hybrid. So let’s just add red, green and purple. I didn’t know what the answer was at the beginning. So I said, how different are these things? If that is truly the way to organize things for consumers, then the purples, the reds, and the greens would all be in their respective places and that’s not what you see. They’re mostly just completely intermingled.

So that labeling system is not capturing the structure of the data. But if you just know the dominant cannabinoid and the dominant terpene, that tells you a lot, not everything but a lot about the product. So notice that these are THC strains, so there’s way more of them than CBD strains. But then notice the color is just telling you the dominant terpene. Lemonene, Myrcene, Terpinolene. But remember the location of the dot is the whole profile. If you know the top three terpenes, it tells you even more, not everything, but a hell of a lot. So it happened to be true, but didn’t need to be true that the chemistry of this plant is organized in a way that we were actually able to create a system that was relatively easy.

So that was a long winded way of saying, the basic idea was how do I turn this graph into something people can use and we spend a lot of time or come with a lot of product and design people to figure out how to turn this graph into this stuff. And that basically, long story short, it meant trying and failing a thousand times until we found something that felt right to us and faithful to the data and tested well with consumers. Again, the point isn’t to be scientific. This final visual representation is not a graph. If charts and graphs worked, we wouldn’t have this problem. We wanted to build something that allowed them to quickly and organically see what the chemistry of the plants is without having to look at numbers or pronounce like really difficult words and this is what we got to.

MRR:                                 Are you hoping in the future to be able to create, say, customer profiles on Leafly to be able to tie with your brand partners so you can suggest things that can match with them?

Nick Jikomes:                   Yes. So the direction Leafly is always going to be moving towards is greater and greater personalization. If you think about it that’s necessary for any technology company today, but it’s especially necessary in cannabis because it is such a personal experience. We all have our own unique endocannabinoid system, our own biology, so we each act differently to these products. These products themselves are complex and nuanced, and so what I want people to be able to do on Leafly before too long is log on and be able to get personalized recommendations. So you want this type of product that looks this way for you. It may or may not be the same product we recommend to someone else.

So if you think about, you know, Netflix or Amazon, right? When you log on to Netflix, and I log on to Netflix, we don’t see the same thing. They’ve tailored that experience to you based on what you watch and what you like, and it’s going to be very similar on Leafly as we move forward. So we’re not just going to stop at blue dream generically, it’s going to be Blue Dream by this brand, in this region nearby to you. Then it’s going to be, ‘because you’ve liked this thing and you haven’t liked this thing, maybe you try this other particular Blue Dream or maybe you don’t try Blue Dream you try something else’, and it’s going to be tailored to the individual.

MRR:                                 Now touching on that, but from the brand perspective, are you thinking about possibly using this information to tell brands, “Hey, there is a gap in the market here”?

Nick Jikomes:                   Yes. Yeah, I would love to. We don’t have something like that today, but this is a framework that is going to allow us to do that, and I’m very interested in doing that because there are so many gaps in the market. What we can do is, so imagine people come on to Leafly and they start building the strains that we can record for them. So let’s just say people start building a particular strain. It’s like, oh, for some reason people like these purpley strains with like some diamonds and some circles, maybe that strains not out there with a lot of people are looking for that we can now start to tell that to brands. It’s like, “Hey there’s a gap in the market”. There’s a lot of latent consumer demand and if you start making that product, you can meet that demand. There’s just no map for people to use to do that today and so we’re very interested to build tools like that for brands.

MRR:                                 What’s your international strategy looking like?

Nick Jikomes:                   Yeah, that’s a great question. So we have offices in Toronto. We have an office in Berlin. But the rest of the officers are all in the US I think. We actually just hired a VP of International to help define the strategy there. But to give you one example, Germany’s medical market is very medically oriented. It’s a very different culture than here in the United States. So next spring for example, we’re doing a medical conference. I’m going to be there and we’re going to be hosting some of the top physicians and researchers in medical cannabis to give talks. There’s going to be a lot of physicians present and so the type of content, the type of information we put out, is going to be somewhat different for that audience given you the difference in the audience. We’re going to dive even deep more deeply into the data. You can imagine for a physician or a strongly medically oriented audience, they really wanted to get to the data even more so than the recreational audience. So that would be one example of our international strategy.

MRR:                   Here in California, you’re partnered with SC labs. For other established states, you have your other partner labs. For new emerging markets, how do you acquire new labs for the partnership program?

Nick Jikomes:                   It’s really one at a time. So we do a lot of proactive reach out to the labs and we say, “Hey, we’re doing some exciting stuff. We would love to potentially partner with you.” but we have to go through a lab evaluation process and that’s easier said than done because everyone’s busy. There’s a lot of labs out there and so it just takes time. We are also trying to work with brands so that they can refer the labs they use to us so that we can go through the process that way. We don’t have a lab partner in every state yet because it’s really difficult to find labs that are doing good science. We want to have lab partners everywhere, but for the foreseeable future we’re going to be in most recreational markets and some medical markets, but others we are not going to be in yet. We don’t have a partner in Arizona yet. We don’t have a partner in Massachusetts yet either so that’s a huge priority for us. We would love to get referrals if we can.

MRR:                   For retailers or cannabis brands that maybe want to partner, how would they be able to team up with Leafly?

Nick Jikomes:                   So if you just contact us we are very excited to work with as many brands and retailers as possible. Leafly already has a very large network of retailers and brands that we work with. So we’ve obviously been doing a lot of communication with them. We would like to continue working with them as well as new brands that haven’t worked with us before. But as I mentioned, we’ve got a lot of room to experiment here and a lot of things that we think we can do right out of the gate, both in terms of digital and physical applications. So we would love to have brands showcase themselves using the new system on Leafly.com and the app in order to physically bring this to life in the retail environment. I think if we can help connect the dots for the consumer so there’s a seamless experience from the digital and the physical environment, that’s going to help create consistency, which is going to help create better education for the consumer, and that’s gonna that’s gonna be good for everyone.

You can find out more about the Leafly visual guide by clicking here. Make sure to get your brand listed by clicking here, or your retail dispensary by clicking here. As always thanks again to Nick Jikomes for his time and thanks to the entire Leafly team for their hard work.

Stay up-to-date!

Get all the news and info straight into your inbox that you need to help grow your business.

To Top