Full episode transcript below. Beware of typos!
Nick Jikomes
Kevin McKernan, thank you for joining me.
Kevin McKernan 3:28
Thanks for having me. I really appreciate that we so we obviously got to know each other on Twitter and we're met with Kevin Jacobson, but it was a it was great to have you reached out and decided to chat about this stuff.
Nick Jikomes 3:40
Yeah, so can you start off by just giving everyone a sense of who you are and what your background and sciences?
Kevin McKernan 3:46
Yeah, so my background began on the Human Genome Project in Cambridge back in was that 9596. So I was the started, there's just someone in the research and development team there and eventually, a lot of those folks left and I ended up inheriting, running a team which I was wholly unqualified to do at that time, I was fresh out of college. But attrition kind of buoyed me up and I ended up managing what ended up being a 10 or 12 person research and development team making robotics for the Human Genome Project and building out their DNA purification tools that they could use for sequencing DNA with this is under Eric Landers. lab down at the Whitehead Institute. I wrapped up that around 2000 The human genome project finished we decided to spin some of those purification technologies out into a company called Agincourt biosciences, which is up here in Beverly mass that ended up selling a lot of these DNA purification kits you may be familiar with in the field known as spry. I think they're under Beckman. Coulter is now label of they have an RNA I think there's a couple different names for them now but a spy is a core technology there that is an magnetic bead based tool that can capture DNA from our from viruses and from from plasmids and various genomes is fairly universally used in the next generation sequencing space. But while we were developing that, Beckman Coulter came to acquire the company didn't know what to do with one of our projects made a project that was trying to get sequencing off of an individual's single micron bead. At the time, we were using millions of these beads to harvest DNA and RNA from organisms. But we weren't really sequencing off of individual beads. We were using them sort of in Samba to captured large amounts of DNA out of organisms, but we did figure out a way to sequence DNA right off of those individual beads. And that turned into the solid sequencer, which applied biosystems came to acquire that branch the company a year later, and then I went with applied biosystems for five years to bring the solid sequencer to market, I worked on the einhorn platform as well, we decided to acquire that about five years into it, four years into it and manage the research and development team. They're mostly focused on emotion PCR, to feed the sequencers, both the solid and the Ion Torrent relied on these types of piece single molecule PCR tools for sequencing. And then after that, I decided to get I had a non compete with the company was now life tech, they merged and became this huge company now they're part of thermo, even bigger. And I had a non compete with them. And I really couldn't do much in the life sciences space. So I decided to go off in an area they weren't in, which was in cannabis genomics. So I left and started medicinal genomics in 2011. And we just dipped our toe in the water by sequencing Chem dog and Le confidential from DNA genetics and put, put those things public. And that's how I got to know John page, a lot of other folks who they'd actually been working on that problem longer than I had. And they published really nice paper that had some RNA sequencing, as well as sequencing from purple Kush and Finola. And that kind of started medicinal genomics, and we got focused on you know, where can we deploy genomics in the cannabis space and make a difference. And so we began focusing on doing DNA sequencing for people building PCR tests that could look for certain genetics in a plant. And then also looking at microbes and viruses and the plant that may infect effector yield. There was a little stint in between there were medicinal genomics was housed inside of the clinical das diagnostic company, which is where I learned about all the the clinical requirements for doing PCR in a clinical setting, many of which actually are sadly being ignored right now in the COVID pandemic due to due to urgency, but they're slowly getting reinstated. But yeah, that's that's kind of sums up where where I started and how I ended up here.
Nick Jikomes 7:31
Interesting. So So you have a background in genomics and the related biotechnology. So you ended up starting a company, which basically does genomic stuff, as a service for the cannabis industry, the legal cannabis industry. And you mentioned, you mentioned some interesting names there that many people maybe won't recognize Chem dog, and some others, those were cannabis strains. So what can you tell us about the cannabis genome at a very high level and what it looks like and how different it actually can be between these different types of cannabis?
Kevin McKernan 8:07
Yeah, so that's something we didn't fully appreciate, in 2011, is how repetitive it was, the genomes wouldn't assemble very well, back then we only had Illumina technology, which is really short read tools back then, I mean, PacBio was just getting going. But so these are 100, these can read about 150 bases on two ends of a molecule. So maybe 300 bases in total, the genome, the first draft of the genome, we put together only assembled into like 400 Mega bases, and we now know it's twice that size. And a lot of that is because of that, you know, over 70% of the genome is highly repetitive, and the short reads can't sort them out. So when you go into assembly, they pile all of these repeat reads on top of each other when they're, in fact, they represent different regions of the genome. The other thing we've seen over time is, I think back then we were assuming the polymorphism, right, that we could do that we could tack in the genome was maybe a variant every 100 basis. And that was largely an artifact because we didn't have the genome is nicely mapped out as we do now. Now we know it's closer, if you compare hemp lines to, let's say, type one lines, you can see almost a variant every 50 to 70 basis. So it's far more polymorphic than we originally estimated. It's a bigger genome that were originally estimated. We also have now genomes of males that were done in PacBio. And those show that the the polymorphism rate on the Y chromosome is like three times higher than it is genome wide. So so that's a real landmine, on the on the on the Y.
Nick Jikomes 9:31
So So to summarize, so far, the cannabis genome is larger and more complicated that maybe people appreciated initially, you said it was 800 Mega bases, so that means 800 million letters of DNA. How does that compare
Kevin McKernan 9:48
the 850? Actually, if you bring in the males, it's bigger. The male has the Y chromosome, which is like 110 Mega bases. So it's the biggest chromosome in genome. It shares about 30 megabases with the X chromosome on on, so females are xx and males are xy. So there's the the male genomes are a little bit bigger. But I'm sorry, yeah, you had another question I cut you off.
Nick Jikomes 10:12
So how does the size of that relate to say the human genome just to give people a relative sense?
Kevin McKernan 10:17
Oh, that's a good call. So the human genome is, like 2.9 gig as a haploid. State, so double that when it's deployed. But we tend to condense diploid genomes of organisms that don't have a lot of differences between them with a fairly inbred and came to a bottleneck, like the human population, they tend to distill them down to the haploid size. So you'll typically hear a number thrown around about three gig for the human genome. Reality is you actually have six gig in every cell, just that your mother and father genomes aren't so different that people call it three. And cannabis, I don't think it's fair to do can because the the mothers and fathers can be so distantly related and have so many copy number changes between them that we probably have to go back to using diploid genomes. This is something that we've kind of struggled with, with a lot of the cannabis references that have been built is that we have this tendency, the assemblers do this not It's not any intent out of the out of the researchers, but the DNA Assembly algorithms tend to condense things into a single reference genome, as opposed to recognizing that the organism is deployed. And you should really try to peel apart the maternal and the paternal genomes into separate genomes. We have tools now that we didn't have back then that can do this, there are these long read sequencers, from PacBio that are very accurate, they have a Hi Fi chemistry now that has a better than saying or sequencing quality. And those 20,000 base pair reads like this. So we can now put together these plant genomes, and split the haplotypes, if you will, into different buckets. And that's, that's ongoing work that's still happening, I think you'll see the references we have up in NCBI. And that the other group who did CBDR accident, MCI, they've all been distilled into a haploid representation of 10 chromosomes, we're hoping to go is to convince NCBI or others to maybe split those into 20 chromosomes because it's very difficult for us to make the assumptions of which one of the maternal or paternal chromosomes should be represented in the genome. So like the probably the most used one in there is a female genome that's a CBD genome CBDR x, because it was one of the earliest ones put in, we put in a lot of these aluminum ones, but they weren't referenced grade, you didn't have chromosome assignments, we've, we've gone through the Jamaican line genomes and done a trio where we sequenced a mother, a father and the offspring, six of the offspring. And that's given us a lot more confidence in what are the maternal and paternal haplogroups because we can kind of follow the inheritance of these things into the offspring. And we've just now organized that into chromosomes with we've had it organized in chromosomes for a while, but we're just we have to re annotate it to get into NCBI. When that process right now, but you what you'll see up there is currently a Jamaican line assembly, which is very good continuity, which is really packed BIOS trophy, not ours, they we just use their technology, and it ended up putting it into over almost five mega base and 15, the average contract size the genomes about 5 million bases long. That's a good like five fold more contiguous than what's in there currently with CBD RX, but we don't have it named as chromosomes, which we we have here internally, we're just trying to get public. But that is a it's a very complementary tool, because these genomes we're finding are so different, that this concept of having a single reference genome that we can get away with and human genetics doesn't really apply in plants, we have to really consider multiple reference genomes, when we're looking at these plants, because they're there, they're almost they're different enough that it matters.
Nick Jikomes 13:40
Yeah. So So I guess the, the punchline here is that the cannabis genome, if you were to take two random cannabis plants and sequence their genomes, they're going to tend to be more different from each other than if you were to take two random human beings. Then you briefly mentioned that yeah, if you you mentioned briefly that, you know, the reason for this is, for those that don't know, human genomes, person to person are relatively homogenous, meaning that you know, even if two people look quite different, their genomes are still very, very similar. And that's because there was this huge population bottleneck at one point in in our lineage. And we're effectively all several billion of us on the planet are descended from something like effectively 10,000 people. Is that accurate? Yeah.
Kevin McKernan 14:21
Right. And there's only so there's only very like one every 1000 letters, cannabis. It's like, it could be one every 50 letters. I see. So the MC noses first Yeah, this is kind of I think the mosquito lineages you can find like a variant every 20 or 25. basis. Oh, wow.
Nick Jikomes 14:36
So so a lot of these organisms are just more genetically diverse than human beings are. And so that makes life more interesting, but more difficult for someone like you in different ways. You mentioned that there's a lot of repetition in the cannabis genome, where does where does that come from? What does that mean?
Kevin McKernan 14:52
So that's mostly from these retro transposon elements. So some of these genomes have these viral elements that just replicate the themselves and copy themselves all over the place. And this is certainly happening in the cannabis genome, there are a lot of these ltrs that are long terminal repeats and Copiah transpose ons. These are things that basically code for a gene that can excise the DNA, I should say make RNA out of a certain segment, and that RNA codes for a polymerase and an enzyme that will copy it back into DNA and put it back in your genome. It's almost like a, like an HIV part of the genome, if you will, it's got a piece of like, got a gag in a pole region, which are things that can polymerize its own stretch of, of RNA into DNA, and then re insert it back into the genome. So it's like a retro transposable element that then kind of copies itself all over the place. And it tends to do this in certain regions of the genome more than others. Now that the cannabinoids synthase genes happen to be buried in a forest of these things, these these ltrs. And that's been one reason why it's been so difficult to like resolve the BT BD allele over time, this is an allele that at the end Domeier described that's that governs cannabinoid synthesis in the plant. It's buried in this huge ltr forest, believed to be more centromeric than telomere a lot of the genes in the cannabis genome are out on the ends of the chromosomes. But the cannabinoids are odd, and that they seem to be more central American buried in some of these repeats. But that may also be what's accounting for some of their their inheritance patterns, these things may have replicated the cannabinoid synthesis pathway into more than one gene. So we now know that there's THC synthase, which is a different gene than THC synthase. We also know there's about five to eight copies of cannabichromene synthase, and a lot of these cannabis plants, and then there's this this, this lineage of all these other cannabinoid synthase genes that well, we don't really know if they're active yet. There might be some RNA transcription off of them. But we don't know if that RNA gets fully processed and put into into proteins that make enzymes are those which are responsible for making the wide diversity of other cannabinoids in the plant. We don't yet know that yet. All those genes need to be cloned and expressed and fed precursors to see if they make these other rare cannabinoids. Like maybe can, can have a mo van is one that came up recently. And I think tch P was found in car Magnolia in Italy, like what is what are making those things we we don't know those are different synthase genes or if those are like differences in upstream genes that might be messing with the propyl side chains versus the heptyl side chains, the input to the synthesis pathway could be altered as well. Interesting. Yeah. I think they're playing a role in the cannabinoid diversity that the plant seen. And I do think there's been bottlenecks, legal bottlenecks that have been occurred. So we've had prohibition that's pushed everything below point 3% thc. So hemp lines have now been breads, just eliminate certain cannabinoids synthase genes. And then of course, prohibition had this inverse underground effect where people bred for very high THC content because they were persecuted based on weight. So now we have you know, skunk, like genetics that are really effective at folding THC from precursor. So it's interesting in that I think that's that's created some convergent evolution in the field, we just put out a little preprint describing some of this that you can find two different ways in the genome to blow apart THC synthase. Typically, it's a copy number variation where the entire gene is just missing. That's probably 95% of the genetics we see in the CBD lines is they they don't have THC synthase the entire genes gone. And there's some rare type for genetics out there which make CBG which is the precursor to CBD and THC. They can't have a CBD gene or THC gene, or those students need to be broken so they don't flip CBG into any THC or CBD. We found some plants that just have both THC and CBD deleted. And there's also plants that still have a THC synthase gene, but there's two point mutations in the gene that are homozygous and change. One of them changes. I think it's residue 355 to an N which is disparaging, which can be quite oscillated. And there's all this literature from xar pal on when you like hostile a certain residues in this THC synthase you alter its activity. So we think that's what might be going on and all those type for genetics. But what's really interesting is that the plant has evolved multiple ways to knock out THC synthase. And that that was kind of a surprise to us. We hadn't recognized that until we got about 1500 genomes deep into this.
Nick Jikomes 19:19
I see. So So I guess the the rough way of describing this is there are multiple genes that encode proteins involved in the production of cannabinoids, THC, CBD, a variety of other ones that we can get into in a little bit. A lot of these genes are buried in these forests of repetitive DNA that come from these interesting genomic entities that you described. And they are tending to be near one side of the chromosome and not the other. And so over time as people bred different types of cannabis for different reasons. You're saying that on the one hand, the black market where a lot of the high THC can Cannabis was bred in the US and elsewhere. Basically, the plants were bred and selected such that only the THC producing genes were were retained and some of the other genes were either inactivated, or or changed somehow so that you're not producing some of those other minor cannabinoids, whereas with the hemp lineages that produce a lot of CBD, there were a couple different pathways by which the breeders were able to create plants that didn't have some of the genes relevant for producing THC.
Kevin McKernan 20:31
Right, exactly. So there's either breeding for, you know, their final end result was just probably measuring with a bioassay, either consuming it themselves, or if they're, you know, later years, they may have been more HPLC driven, but they could just look for THC compound as a percentage of flour. But there was more than one way to skin that cat, you could you could be selecting plants that had complete deletions of the THC gene, or ones that had point mutations that inactivate it. So that's it, I guess it goes to show that there's multiple different ways to get to the desired end. And I think that's a theme that we consistently see in cannabis, perhaps more so than in human biology, because it's so variable, that when you go and perform a selection, you can probably find multiple different variants that achieve the same end. And I think that's I think we have now have evidence for that on at least for the type for plants.
Nick Jikomes 21:22
So when you when you say type 1234 These are referring to plants that have different genetic such they produce different cannabinoid profiles. So a type one plant would be a THC dominant plant that is producing mainly THC and not some of the other cannabinoids. Those plants are the most prominent the ones that you're most likely to encounter in, in a dispensary say,
Kevin McKernan 21:44
yeah, yeah. So
Nick Jikomes 21:45
one of the things a lot of people notice when they go into a dispensary is there's so many different types of type one or THC dominant strains out there allegedly. And so you know, you mentioned some earlier Chem dog and some others. So if you walk into a dispensary, and you're seeing Blue Dream and chem dog and OG Kush, and this and that, how real are those do? Do most of those things map to genetically distinct cultivars?
Kevin McKernan 22:12
So that's a good question. We so we've so good, if you go into canopy media and just pull up Blue Dream, which I might go to do live here, you'll see a lot of them are in fact clonally. They're clonally. replicated, which isn't always the case. Let me see if I can pull up blue dream on here to get the space name so Oh, I should probably pop this up on something you guys can view as well. I can probably share a screen if
Nick Jikomes 22:46
you if you can. Screen Share.
Kevin McKernan 22:50
Okay, so let me hit screen share. This is just a view of Blue Dream happens to be one that we sequenced a bunch because the people were asking the question, how do we know it's there, there was some multi state operators that want to Blue Dream in every state, but you can't cross state lines, they got to grow differently in every state. And there was concern over over those genetics, whether they were all the same. And let me share the screen, it might be on your end, I may need a approval to do so. There you go. Yes, saying it's disabled on your front. See if you can do it now. All right, there we go. Let's go here. And share. Alright, so this is a website called canopy that holds a lot of the genetics that are public for sequencing. So if you just go into the search bar up here and search for Blue Dream, it'll give you a list of 37 strains that are in the database that have blue dream in the name. And so you'll find some that are, you're probably you probably won't find Snoop stream in here. But there's a snoop stream that looks a lot like this blue dream. So that here's one from Happy Valley, which is sold here and a dispensary in Massachusetts. So these charts here, we look at the x and y ratio, the DNA to figure out if it's male or female, usually people are sending females. And so that's not a surprise. We look at the hetero zygosity, which is gives you a sense of how inbred it might be. That's helpful for people making seeds, they want to breed things. So such that the heterozygosity hopefully is low, so the seeds don't come out very different. But this is kind of a predictor of how different your siblings will be if you if you would have to breed it. We look at the sequence coverage over four different cannabinoid synthesis regions because we find this is very predictive of the type of plants going to be in this case. This This one has both the THC synthase gene but this over here, they're really the depth of coverage here how it's kind of square wave over here. This is really a sign that there's not a lot of depth of coverage over CBD synthase and it's gone, as opposed to these other ones have like 50x or 100x coverage over these other genes. And this one has a CBC cassette this is a this is a two beta two megabase deletion that often occurs in cannabis plants that we just track because we were suspect that can have a chroming genes may produce Small amounts of THC and play a role in whether you get a point three or point 4% thc plant, that's still a bit of a theory. It's not not been proven, though, because epi medelita has done some nice RNA sequencing work on these genes to show that when this is around, there's a little bit more residual THC. So this just tells you there's a variant in this gene that's this has been described as this particular variant, we should probably get a link in here that to the paper, but this has been described before by no free. So this is a known variant in THC synthase, that shows up don't really know if it does anything to it yet. These are a bunch of other variants in that pathway. But to your point of how similar are they? This is a tool we use to look at the genetic distance between other things and canopy media. So you can see all of these blue dreams lineup, and they even are clonal to snip stream, which someone probably just renamed Snoop stream to Blue drown. So a little better.
Nick Jikomes 25:48
Yeah, we definitely see that. I mean, in the profiles, the snip stream, the Blue Dream profiles are remarkably consistent across almost every producer.
Kevin McKernan 25:56
Yeah, so this would be fantastic data intersect with what you've done. Yeah, it's the database is a little weak on chemo types. There's a, you can search by chemo type, but there's just not a lot of chemo type information that gets piled in here. But we can do it if people send us their their se lab report or whatever lab they're using, we can we can scrape that information and sort it with their restraints. This particular color code here, these numbers are more important to color code, we kind of someone in our group just set this at 10%. And that's not necessarily a hard distinction between clones, I think this is going to get refined back down to more like 5% over time as we've sequenced more and more clones and more and more siblings. But you can see some other things that might be you know, related. And at the same time, we try to search the database for things that are most distant, and you'll see things like cherry blossom, which are CBD lines pop up as being, you know, very distant to Blue Dream.
Nick Jikomes 26:48
Okay, so but the point is, you you're basically you're sequencing the DNA of different plants. And it's quite simple for you to tell whether or not samples are closely related genetically. So someone has blue dream, someone else has snoops dream, someone has this other strain called Serious happiness. And they're very, very, very similar genetically. And you can see that,
Kevin McKernan 27:10
we can see that and we what we have tried to do so that the data can be more correlative to some of the work that you're doing is we try to cover the genes in the cannabinoid synthesis pathway and the terpene synthesis pathway. When we do this, we don't have, you know, great variant to terpene predictions, I mean, Felipe, Henry's done some work on this, which has been really, really good. And then he's got a couple snips that kind of predicted broad class of terpenes that you may make, like you might be limiting, dominant or peddling dominant, but we can't tell you the the, oh, you're going to be 3%, beta caryophyllene and 1% accumulation type of thing yet. But I think over time, as more people correlate these variants with kind of the, the outcome will get close, I don't think we're ever going to nail it perfectly, because we're just predicting the hand that you're dealt, not, not how you play it, right, like the conditions upon which you grow these is gonna obviously alter the terpene expression and how you cure it. And a lot of those things are gonna are going to alter the perhaps the magnitude, but there are some genetic switches that really determine how much you can actually make like a plant that doesn't have a THC synthesis synthesis synthase gene, let me see if I can actually go to one like cherry blossom probably doesn't have on, it's not going to make a lot of CBD, we can predict that pretty readily like here's, here's an example where the CBD gene is present. THC is gone. There's a neighboring gene next to THC that looks a lot like CBD synthase. It's often deleted, we just we happen to watch them both. But they usually they're both co deleted. Usually, it has cannabichromene. So this might make CBD CBD alone when it's limited to yeast zurple shown that it'll make a 20 to one ratio of c CBD to THC. So it's not that this gene itself is a little dirty, it makes a little side product, we suspect the same thing is going on with cannabichromene synthase that it's a little dirty as well and sometimes makes a small amount of THC in the process of making cannabichromene Not as well proven as it is with what they've done with CBD synthase. They've actually cloned this and shown that that happens. But nevertheless, we can predict that if you don't have a THC synthase gene here, you're not going to be making a five or 10 percenter, you this is really about dissecting whether you're at a point three or point 5%, which the genetics is getting close, but not perfect I'm doing yet. So I don't think we can maybe if the if the settings were put to 1% We could probably predict genetically where you're going to fall. But I think splitting hairs between point three and point four still very difficult to do with the DNA alone. So
Nick Jikomes 29:29
so if we back up for a second go to the top, so go down to like the heterozygosity chart. So so if you're a cannabis cultivator you're making cannabis plants you're breeding them. Your goal is to produce a nice, consistent set of strains that will actually be sold in dispensaries. There's a number of things you might want to do in terms of genetic testing, and that's where you guys come in. One is you want to know if some seed that you have is male or female so they can send you a sample. And you guys can tell if it's male or female by looking at the ratio of X chromosomes to Y chromosomes present in that.
Kevin McKernan 30:08
Yes, yeah, we've actually turned this into a PCR test. Now, we tend to sell that to a lot of labs who do it locally because sending some of the seeds around can be complicated on the farm bills helped that out on a on a hemp side. But we just kind of made a decision, we're not going to open up a service lab here and compete with a lot of the other labs that our customers we just sell the picks and shovels to them. But what we do do is we run some validation studies that help guide them on how to utilize the tool. So for instance, with the seeds, people don't want to test a single seed at a time, they want to test 100 seeds in a batch of 1000 or 10 1000s. And know if the batch they're about to buy has any male contamination in it. So we set up a study that looked at taking 50 Female seeds and 50 Male, female and male seeds. We did this with Colorado seeds actually, some Jesse helped us out over there. And they basically goddess ratios of seeds that were mixed at different ratios. So we had where it's 5050 male female seeds, where it's 7525 and all the way down to 100 seeds 100 to one, we can pick up a single male seed in 100 seeds with a qPCR test that just shows the that the Y chromosome is present at a at a certain Ct value that tells us there's some male DNA sitting around. And
Nick Jikomes 31:22
for those not familiar with the cannabis industry, why would a cultivator want to ensure that there is no male contamination?
Kevin McKernan 31:31
So well there's a couple interesting developments there. So traditionally, they don't want male seeds around because of a male pops outdoor and a large grow. It can then pollinate all of the females they seed out which makes lower quality flour lower cannabinoid content flour and just creates a wreck. Now Oregon CBD has done some great work making triploid switch don't make seeds. So that kind of changes the game quite a bit. And and they they're mostly focused on CBD and CBG plants but you know that technology can easily port to THC or type one plants. I think dark heart just announced that they've done something on that front. But that's a path of making sterilized plants which can be very helpful for planning a very large grow that you don't ever want to get pollinated, particularly in a place like Oregon where there's a lot of back you know, backyard grows that could pollinate your CBD farm with THC and then suddenly, the next generation of your farm may have some seeds that sprinkled on to the ground that pop hot and if you get inspected and you have a plant that's above point, 3% THC, they can come and burn down your whole your whole garden so it's really important out there and these outdoor grows because pollen, cannabis pollen can travel miles. So someone could be growing in Willamette Valley, you know, down down down the valley a mile away and blow pollen onto your CBD garden or farm and really record for the following season. If they don't, don't clear that farm with any seed that may have had THC pollen blown into it. So they likewise when people are buying the seeds to plant they want to make sure they don't have any males coming through. Because if one male sits out in a in a in a farm of 1000 females and pollinates them all you can wreck you know an enormous crop by having those males float around. So they want to buy certified seeds that don't have any males coming through. They're fully feminized was in the feminization process never 100% I mean, I think the numbers I saw from Oregon CBD is that they have like one male every 4400 plants, which is really remarkable. But there's still one that can be a male in that whole feminization process. And that that's the that's what they have to kind of keep their eye out for in it as they're planning these things. Others that are selling these, these seeds, a buyer wants to know like, what's that ratio like, test 100 seeds a couple times and tell me if there's any y in there. And if there's you know, three lots come through 100 seeds that are clean, then we suspect the whole batch is probably clean and folks can can move on so that I think really doing that ensemble, a seed test is going to be really, really important for those that are trying to sell these, these lots of seeds.
Nick Jikomes 34:04
So the other thing that I want to ask you here is you're showing us some of this data to do with these cannabinoids synthase genes, sometimes they're present. Sometimes they're absent. You've described for us how, you know in certain lineage lineages of cannabis, there couldn't be deletions where a gene is missing. You might have mutations that inactivate a cannabinoid synthase gene and things like this. So thinking about the ancestral sort of wild type cannabis that was out there, did that cannabis used to have all of the the cannabinoid synthesis and produce some amount of most of these cannabinoids or how do you think about the ancestral state of the cannabis genome?
Kevin McKernan 34:43
So my grant, I've not been in the field that long. So I'm, I'm speaking from what I've read and what I've heard. I suspect the prohibition created a cannabinoid synthase bottleneck where people were breeding in two opposite directions. They're breeding toward hemp lines that had no THC and then they're breeding toward plants. have very high levels of THC, because those were, if you get caught with a pound you want that pound to be 20% not 10%. Right? There's just like whiskey during Prohibition right there's there's a tendency to concentrate the product when it gets prosecuted based on weight. So I think that that may have selected for plants that have a particular cannabinoids synthase ratio, if you will, or types of genes that promote those types of cannabinoids that we're measuring, because we're not constantly measuring for THC P, we've probably bred it out of the population to very low levels. In fact, it's no surprise that they found that in the Italian car Magnolia line, that's a line that's really a CBD line, or fiber line, right? It's not known to be a THC dominant plant. So I think there's a lot to be done going back and sequencing a lot of the land races to try and find some of these extinct cannabinoid genes are genes that just aren't in the commercial dispensary market that are getting propagated there. There's probably a lot of exploring to do in some of these, these older varietals that may not have been hyper bred and one cannabinoid direction or the other, then the reason I, I say that is that these things are all competing for precursor. So when you when you when you drive a plant or making, you know solely THC, it's gonna come at a cost of making the other cannabinoids. It's rare that you find a type two plant that's making 20% of both CBD and THC, it's usually making 10 and 10 of each, right. And that's because it's limited on the amount of precursor that you can feed into the pathway. And since they're competing for those resources, they can't make the whole library of them. So maybe some of the earlier land races that were less perhaps less potent on the THC and CBD front might have higher concentrations of the other cannabinoids. I don't know if that's that's just a theory of mine that I suspect to be true. If we went back and survey many of these other land races that aren't they haven't necessarily been bred for fiber production or CBD production or THC production, we might start to find some of these rare cannabinoids.
Nick Jikomes 36:58
Interesting. And then you mentioned that there are now cultivars out there strains of cannabis that have for example, high CBG. What did those look like genomic Lee and how common are those at this point.
Kevin McKernan 37:11
Those are somewhat rare. There's one here in front of you, if the screenshare is still on, you can see one called double black label that just came through sequencing fairly recently. And this one is got a very interesting mutation. You don't let me I'll find a let me pop onto another one to make sure I think this is one as well. It's the CBG right in the name. Just so everyone's convinced it is what you'll see here is they don't have any CBD synthase gene. So they they're not making CBD, but the THC gene is there. So that was a real odd to us. That's what kind of went Put us down this road of looking at these things a little bit more deeply. So what you can do in Canada pedia is you can see that over here is the variant frequency in the population. So one of these variants, the searing 355, disparaging this thing is only at 1.8% population frequency in the database. This one we see all the time, this pro three three arginine and THC synthase. It's in a lot of type one plants. So it's like this probably isn't the damaging mutation that's taking it out because we have really high THC rich plants that have it. And it's great 10% of the population, right? But this very rare one, actually will. You can see if you actually go into this through this little IGV portal we have whether it is this one here is actually a homozygous mutation. That means it's an amino acid change that is in both the maternal and the paternal genome for this plant. And we believe this is what knocks it out because this is common to a lot of the CBG plants we see in the database is they have this particular mutation that is in THC synthase and this is right in the zone of the gene that sir Powell has heavily mutagenic he went through mutagenesis and changed a lot he's immediate, you know, as is not this exact one but a lot of ones in the neighborhood. And that For those not familiar this particular this is called like an s 355 n mutation and when this when when things mutate to ends, you start thinking about like constellation because those those amino acids constellated. And when they get glycosylated. What zurple showed is that that alters the efficiency of the enzyme sometimes when you like oscillate, it gives you more THC and sometimes it knocks it knocks it down. In this case, we suspect it's knocking it down. It's only being found in THC or I'm sorry, in CPG lines
Nick Jikomes 39:35
I see so So the moral of the story here is this is a plant this is a type of cannabis plant that produces high levels of CBG in the end, and the way that plant is doing that is that it has on the one hand gotten rid of somehow the CBD synthase gene so it has no potential at all to make CBD basically. And then the THC synthesis gene is there but it has acquired me mutation that probably renders it really bad at turning the CBG acid into THC acid. Yes,
Kevin McKernan 40:05
that's that's that's going theory. And what's really interesting as we found one sample in the database, let me see if I can pull it up Doug's Karen's in here, right? Everyone knows everyone may know Doug's Baron is being a thc v line, right?
Nick Jikomes 40:17
Oh, yes. That's that's the one from California. Yes,
Kevin McKernan 40:21
yes. Yeah, yeah. Oh, I'm missing an S, get the database. Do that in the search, but do There we go. So what's interesting about dogs Baron, is it has the same gene mutations. So initially, when we saw this with this destroys the whole theory, because this thing makes THC and THC. But what's interesting about Doug's Varian is it's heterozygous at this position. So it's got one copy that works. And one that doesn't. And what's really interesting about the the cannabinoid profile in this which happened to come on the on the vial, we got this in California, is it was like making half THC, half CPG. So it's got one gene in here that can clearly turn THC over, but probably needs to to really convert all of the CBG into THC. It was making some THC and TCV of ratios. It was, I think it was like 2.6% thc. We have a chemo type and that has preprint we put forward but for those who familiar with THC synthesis, the THC synthase gene has no bearing on whether it's a variant or not. That's something upstream in the process of making the precursors. So THC synthase, if you put it into yeast, if you just change the does a great presentation on the combat from Dimitra acts that did this when they just changed the precursors going in like the hexanoic acid, they would just make that larger carbon chain and you'd end up with either shorter or longer THC molecules at the other end. So they actually showed the synthesis of THC P that one that's like a say a seven carbon chain that's 30 times more potent that was found in Italy and car Magnolia, they actually were able to pull it off in the lab by feeding the yeast different precursors in the TC synthase gene. I have to confirm whether it was TST synthase, or CBD synthase. But similar mechanism would just output different carbon chains off the tail. So we're not surprised by the fact that this is a THCV line but it's making like 10% thc 3% thc v but it comes along along with it is like an incomplete conversion of CBD and THC you get high levels of CBG on the slide because it's carrying this whole this heterozygous mutation or it's got one copy maybe from the mother and one from the Father that are different here. And it's got one gene that's not fully active another that is
Nick Jikomes 42:46
I see. So Doug's Varon is a cultivar that produces that will have in it in the end, THC CBG and THCV. The absolute levels of THC are higher than CBD or thc v but much lower than typical strain that you would find in a dispensary. And it has a few percentage points of both CBG and thc v
Kevin McKernan 43:06
Yeah, let me I should be able to pull this out of a preprint right now. That's, um, so I go for a new window here. Are you I'm sorry, I'll do let me do this window where you can see it. It's opened up the wrong window. But there's a OSF i Oh
wait, no, no, I think this one times No, no, sorry. I preprint servers mixed up. Right here. So this is the there's this the paper describing this particular conundrum of and it's in here. Wow. Seven I wanted Okay, so you're probably not seeing this, are you? Yeah, I can see it. Okay, so there's Doug's veteran right there so that you can see the chemo type on it. So it's making some some keep in mind, it's not fully decarboxylase. So it's coming out as thta CBGa. So it's around four to four and a half percent CBGa and maybe low end and
Nick Jikomes 44:23
I think there's something else you're pointing to that I can't see.
Kevin McKernan 44:25
See if I can. Oh, no, there's way to do that on Sure. Let me show the screen. New shares right here. How's that? There we go. Okay, so you can see the chemo type down here is got a little bit of thc v. And then you got to really combine THC with tfca. In that's maybe 11%. And then the CBG and the CBGa are maybe adding up to almost four and a half 5% Right. I see. We suspect this is incomplete conversion to CBG because the one of the THC genes is broken, like like a CPG line is one floating around that can still trickle some of the THC out. And the fact that it's a Veyron has nothing to do with these these four genes that we're showing you That's something that's still yet to be discovered. We have some hands but there's so few of these plants out there, we only have like three samples in the database that are barren related, I think it's this one and another Black, Black Mamba. Now there's one that we got out of, in Heartland out of a dispensary up there that had that like 7% thc. But we don't have a lot of those samples to like, you know, we have we see mutation and one of the genes upstream pathway, but it's an n of three. So we can't really do anything with that. So we get more Veron samples into confirm if that's the variant that actually confers whether you make a pro bowl or pencil side chain,
Nick Jikomes 45:50
I see and what to I mean, what if anything, do we know in terms of what people say at least the effects of this particular cultivar are given that it's got this pretty unique profile?
Kevin McKernan 46:01
You know, that's interesting. I've always heard that I've not had a lot of experience with these like thc v I've heard like suppresses appetite. But I personally have no have limited experience with it. And then CBG is known to be an n zeolitic And as some similar properties to CBD on that front, but you know, I know of people that take CBG because CBD gives them tinnitus or tinnitus. So it's certainly a different molecule and perhaps provides Superman's oolitic behavior in a different way. But there's there's there's some studies out of prostate cancer too, but they're they're it's much more limited. I'm what's really exciting what's going on in the cannabis field today is that we now have these like full born, you know, CBG cultivars so we can make the CPG oils and we can start to you know, we can start to tease this apart a little bit better, but those weren't really around. I didn't see you know, five years ago CPG lines were hard to find maybe two years ago we started to see like Miriam's hope and a variety of other people, you know provide CBD oils and only today am I finding them in gummy bears like in local dispensaries? Did they have like one to one to one gummy bears that have five milligrams of each THC CBD and CBG? So it seems like it's kind of a new game, which is kind of exciting.
Nick Jikomes 47:19
How, how far are we? Or how difficult is it genomically in terms of how you would breed these things, to make a plant that is you know, THCV dominant that's making you know, the same the same proportion thc v that other plants make THC or any other cannabinoid? Are we right around the corner from that? Or is that something that's gonna take you years to develop?
Kevin McKernan 47:37
Well, that we certainly worked with a few people that have done it already. Oregon CBD, I can't I can't speak to exactly how they did this. They use some of our snip chips to load zero in on a location in their genome that was predictive of cbdv production. And I'm assuming those same genetics, if they were crossed into a type one plant would probably give them a THCV dominant production. But what they did is they they did a study of it probably a couple 100 cultivars, or I should say across they did to track down what would the offspring had the the particular expression and then tried to pin that down to a particular snip from a snip chip that that we sell. And then they converted that into a qPCR assay to track their genetics. And we've been begging them to turn that into a qPCR assay for the rest of the field. But the information have gone back from SAP is that when they've tried to track that, that variant in other populations, it hasn't been as predictive as isn't their own. So there may be this convergent evolution thing going on that front as well, where there you can't bank on one variant dictates everything in in the cannabis population. So he didn't think I'd be very fruitful to run his assay globally, because it wasn't necessarily predictive outside of his lineages, but he was able to utilize that to breed his Ferren lines, they also found that auto flower flower allele and that same study, which again was something we asked if it would be applicable to the broader market, and they didn't think it would be because they weren't finding it to, to predict things outside of their, their family. So what they did is they went through and just carefully selected their their parents and designed the right study to cross them and look at the offspring and quantitate some of these outcomes in the offspring with with other quantitative assays like HPLC, in the case of autoflowers, time flour, and then correlated that with genetic markers, and they were able to zero and markers that could predict how this would behave with their mothers and fathers. But unfortunately, it seems like in cannabis, those studies all have to be done with your with your own mothers and fathers because what we learned from Seth may not always superimpose itself on another genetic background.
Nick Jikomes 49:47
Interesting. So if we let's go out of the screenshare here.
Kevin McKernan 49:51
Yes, yeah. How do you control it?
Nick Jikomes 49:54
I think you do. Just just hit the stop button at the top. I think there we go. Perfect. Perfect. All right. So We've we've covered a lot so far to do with the cannabis genome. One thing that I want to step back and describe for people is, you know, you've mentioned qPCR, and PCR generally. So this is the polymerase chain reaction. This is a very common commonly used very powerful technology that's used to do genomic stuff. So it's gonna, I think, help people understand other things that I want to talk about. But at a very basic level for people that don't have a background in the relevant biology here. What is PCR and how does it work?
Kevin McKernan 50:32
So PCR is a very sensitive tool, it amplifies DNA by copying it with a polymerase and it works in an exponent on like a log two scale, what that means is it doubles every single cycle. So the if you have a single molecule in the reaction, after one cycle of PCR, you'll have to after two cycles, you'll have four, then you'll have a then you'll have 1632. So this quickly explodes into being able to pick up single molecules in within, usually within 40 cycles of PCR. Now that being said, whenever you're using things that amplify to that degree, you have to very, very tight controls, because you can amplify the wrong thing. And if this is the the specificity of PCR is very dependent on the primers that you use to pull out what you want. So for instance, if you if you're not careful, and you design set of primers inside THC synthase, it's very likely that they're going to amplify CBD synthase CBCs, and face many of the other cabinet genes, you have to be very careful putting primers in those genes to make sure they specifically amplify the cannabinoid gene that you're looking for. Otherwise, it'll run off and do all types of harm amplifying the wrong stuff. This is something that I think plagues the whole Coronavirus epidemic. And one reason I was really outspoken about it is I saw what was going on it was like we could never get away with this in the cannabis field. This is just like horrific what they're doing. And it was that they they put PCR assays out there without any controls, and without any guidance on what CT they should be using to call sample positive. So why are the controls really important? Well, PCR gives you a number of this to what CT this exponential amplification crossed a threshold of fluorescence when these PCR reactions amplify, you can do some tricks where you make them fluoresce in the process, and in concordance to how much DNA is synthesized.
Nick Jikomes 52:19
So Kevin, before we get into this, let's try and give people a very simple cartoon example of this. Let's say you and I are both suspects in some crime. And they've got they've got a sample from the crime scene, and they want to see if your DNA versus my DNA is present. How would they do that in a way that makes it unmistakable, that it's one of us and not the other one.
Kevin McKernan 52:43
So that's a good point. The way they currently do this with the FBI is with the CODIS database system, and they will either use str is which is the amplify these short tandem repeats that have variable lengths. And those are multi allelic loci and make use about 13 of those to pin somebody down. But there's a lot more information in an str. And the reason they went with ser is for some of that is because of the technologies in the past were easier at looking at fragments at a different length versus fragments that were the same length, the different base technologies change today, we're now it's easier to look at single nucleotide changes of fragments at the same length with snip chips, what have you, but you need more of them, you arguably need about 100 snips to to really get somebody to have the confidence of you know, one in a billion type of confidence to put somebody to prosecute somebody. So they might look at 100 variants in your in your in your in your genome that will use PCR and then probably some either allele specific PCR method or some other sequencing process to read the differences between those fragments, because in the case of snips, they're usually always the same length assay that you're amplifying, so different than than the str is where you're amplifying and looking for length change in the DNA, whether you can run through gel, here, you're looking at point mutations in the genome. So when you want to do that with PCR, which people aren't doing necessarily in Coronavirus testing, they're just checking presence absence of a virus. But if you want to do allele specific PCR, you've got to be much more careful that you design primers that are very sensitive to amplify one allele versus the other and labeling them differently. 40.
Nick Jikomes 54:17
Yeah, so So what you're saying is, you know, if we're at the crime scene, and we want to find if Kevin's DNA or my DNA is present, we can use PCR to do that, to detect the DNA with very high sensitivity. The problem is, as we mentioned earlier, right, our genomes are highly similar to each other. Yes. So, you know, most of our genes are gonna be exactly the same. So the question is, well, how do you tell us apart, you have to find a gene that's different between me and you, and maybe it's only slightly different. And then you have to use this thing called a primer, which is just like a little piece of DNA that'll stick to that particular region of the genome. And you have to be careful about making sure that primer is specific to the version I have or the version Kevin has, but not both.
Kevin McKernan 54:58
Exactly. Exactly. In this, this is something that has plagued the virus testing front because they're not doing allele specific PCR with the viruses. And these are already viruses that mutate a lot. So we'll keep hearing about new variants that come out, but they screw up a PCR assays like the s gene target fails, you know, that's because they didn't design that to be necessarily allele specific, and not that they should have. But that's just a limitation of them having primers when when snips fall under your primers, they don't have, right, unless you anticipate that variant into your primers. But it just shows that you can't you can do allele specific PCR. And when alleles change PCR sometimes fails. That is certainly something we we had challenges with, when we were designing these sex tests on cannabis. We didn't, we didn't appreciate how polymorphic the Y chromosome. So we kept rolling out assays that worked in one genetic population, were readily able to predict males and females, we go into new population, and suddenly our Y chromosome had a variant under our primers, or acid was failing and calling males female. Right. So we had, we've made four different versions of that test now to dance around a variety of variants that are in the Y chromosome that we keep discovering, and now it's no longer a single test, it's like a test it's multiplex and targets multiple regions of the Y to deal with the fact that the y's are so different. That's a that's a presence absence test that has to be do the opposite of what you're asking. It has to it has to not get tripped up on the fact that we have differences, when you actually want to identify those differences, you have to get a very little specific PCR going or do sequencing of the PCR products, when you're done to know to split apart the differences that might exist between me and you.
Nick Jikomes 56:32
So when PCR testing is used in the context of a COVID test, you get your nose swab, if the virus is inside of you, it'll be contained in that sample. And PCR is going to be good at detecting the presence of that virus, because it's going to amplify up some of the genetic material that's there, can you sort of unpack that for us and help us think about what that means for things like false positive and false negative rates?
Kevin McKernan 56:56
Yeah, so that's a, it's a really important point, and that when you swab your nose, there's a paper up there from da doo da, I think it's da, da, da, o u h, that goes through the amount of variance, you get swabbing your nose. And it's like 1000 to 10,000 fold variants swabbing between people. So that's it, that's a big problem. Because when you want to test for viruses, knowing the amount of virus you have there is kind of a number in space, but you really need is the number of viruses per human cells. So you need the denominator. And some of the first test that came to market didn't have the denominator. And the way you typically measure the denominators, you amplify human gene and viral gene at the same time. And you'll look at the ratios of those CT IC
Nick Jikomes 57:38
IC. So what you're saying is, when you do the nose swab, it's sort of like sticking a spoon up your nose or something. And you're going to scoop out a different amount of virus for each person. And part of that could just be, you know, you got unlucky or lucky with the scoop and got more, what you want to know is how many viral particles per human cells. So you actually need to get both
Kevin McKernan 57:57
numbers. Yeah, you need two numbers. And the test that don't the two numbers are really clinically that would never be allowed in a CLIA laboratory pre 2020. But I think in the, in the panic of the pandemic, people just ran with the first test that came to market. And now, the thing about the combat market is there's like, there's hundreds of different tests in the market. So I don't want to speak broadly about them all, some of them do have internal controls. And that's what and that's what you need is you need something that measures that human denominator. So you can gauge how much viral load you have, when you don't have that denominator, you get in all types of arguments with people as to what the Ct value should be for cut off. Well, if you can vary by 1000. That's like, you know, that's like 10, CTS, you know, you're never gonna, you're never going to agree on what the Ct value,
Nick Jikomes 58:38
what is what is CT, oh, that's
Kevin McKernan 58:40
that's the cycle threshold at which the signal goes positive and PCR. And, and so the later it is, the less DNA you have. And the earlier it is, the more you have, and it's on a lot to scale. So every number represents a doubling. So going from, you go out 3.3 CTS, and that's actually 10 fold differences. And then you go out 6.6 CTS, that's 100 fold, and 10 CTS is basically 1000 fold.
Nick Jikomes 59:03
So what an example here be, let's imagine two people that get a nose swab, and there is some viral particles in both of them. One person is legitimately sick, they have lots of viral particles inside of them. And so there's lots of viral DNA there. And in fact, they are they are sick with COVID, the nose swab will pull out a bunch of viral particles, there will be many particles per human cell. And PCR will detect that quickly. So so very few cycles, we'll be able to see it, the other person could just have, you know, just just use a cartoon example. Maybe there's just one or 10 viral particles that they sniffed up, not enough to make them sick yet, but you just happen to scoop those out. And you can detect that as positive if you do many, many cycles of this test, even though it's not enough viral particles to actually be Yes, causing sickness.
Kevin McKernan 59:51
That's right. So there's a couple of things going on here. There's one the sampling variance in the nose is so wide that you need an internal control to know it. The second issue with this virus that it has a very long lifespan in your body after you're no longer spreading it. So the the papers will show that you might be simply may be infectious for about seven days, but you could be PCR positive for 70 days. And that's because the way this virus infects cells, it gets kind of when it attaches to h2 receptors, it gets kind of vegetated into the cell, and it stays inside this double wall membrane inside the cell. So that virus, even when it's no longer replicating in yourself, your cell needs to die for that thing to go away in your immune system to clear it. And so what we're seeing is this RNA from this virus base, and its unique biology, sits around hides from the immune system by hiding inside your cells inside of like almost its own nucleus inside of the cell. They call these double walled vesicles. And the immune system can't clear it out. But it may it may effectively keep it from from budding from the sell anything bugs out of the seller can whack down does antibodies and T cells have figured it out, but it can say hiding in your cells for months. So we often find people are still PCR positives by the CDC has like a moratorium on PCR 90 days out like don't do it again after 90 days, because you're probably still be positive it'll be if you won't be sick. Now, that's not really a analytical false positive by PCR. It's finding the RNA and it's really there. But it's a clinical false positive in that it's no longer a quarantine risk, right. I think that's what a PCR went off the rails is they conflated analytical, false positives, clinical false positives, because in the past, we would never just use a single test to make clinical decisions, we would say, Are you symptomatic? And do you have an antigen test that's positive and doctors consider a lot of its information and making a clinical call. But in the urgency to pandemic everyone just said screw it, we're gonna use PCR is the doctor now. And we're gonna call any anyone who's already positive, infectious when they're not in quarantine, you know, too much society to the point we quarantine the nurses need to be in the hospital. So it got a little bit out of hand not handling this kind of long tail of PCR positive that we have with RNA, there are correlates to this that we should learn from in the cannabis industry. When we're testing for pathogens and cannabis. We need to be very careful, we don't call dead pathogens a problem, right. So we've been we've we've designed some tools that try to obliterate any RNA or DNA that's not inside of the cell membrane wouldn't necessarily handle this thing with SARS, because of the unique biology in SARS. But it does handle a lot of cases where you have dead DNA floating around that PCR picks up that doesn't show up on plates. But we see it well we have tools that erase that DNA, because DNA is very susceptible to nucleases that chew it up. And these nucleases can't get inside of a living cell and chew up that DNA. So we have ways of kind of sorting out this live dead problem in the in the microbial testing that we do in the cannabis field. But it's an important lesson is that one, you really got to have these internal controls to know how much how much pathogen Are you measuring in relationship to the cannabis plant? We do that by targeting the cannabis genome with with an internal control? And also, can you skip the part the live dead problem? How many of these pathogens are still living and how much of it is just free circulating DNA from dead from dead organisms? That that stuff you can handle with nucleases.
Nick Jikomes 1:03:14
So we've now touched on COVID testing and we've obviously spent some time talking about the cannabis genome and interesting intersection here that was making rounds in the news recently based on some new studies that are out are the potential antiviral effects. Antiviral meaning anti SARS cov, two effects of some of these cannabinoids in particular, there's a study out now, I actually talked to one of the lead authors on one of the next episodes, showing apparently that CBD, at least at a certain concentration can have antiviral effects for SARS, cov. Two, so can you talk about that result a little bit? What you think about it? Yeah, I
Kevin McKernan 1:03:53
love that paper, actually. And Ethan Russo just did a cast mechanic group here that he has a great a great read and want to listen to this, because he's been through a lot of these clinical trials and understand some of the dosing elements here. But what I gathered from a quick read of that paper was that what I found very fascinating is that oftentimes with CBD studies, we don't have the ability to run these things on patients. So we do them in rats and other animal models. And the dosages are often way too high that we're never going to get to humans, but it did sound like they did some, they made some effort in that paper to go and look at the micromolar concentrations of CBD in the blood of patients who are receiving 1500 milligrams and like Epidiolex and they could see that okay, there were around like, point five or 1.5 micromolar in the blood.
Nick Jikomes 1:04:38
So, so these patients were people taking Epidiolex, which is the prescription CBD you can get for epilepsy, and they're taking something like 1500 milligrams per day. And then the study group not only knew that about them, but they actually went in and looked at what the concentration at that dose of CBD was in their blood.
Kevin McKernan 1:04:57
Right. And that's about it. on their EC 50 curve where they start to see some response. So they're looking at different amounts of CBD that are given over time. Then looking at viral viral load, skin, let me just get some water on the road here. So that's very important because you got to know how much doses you need. And then 15 milligrams is a lot, right? But the, the epilepsy kids manage it, but it's not what you're going to get like at a gas station. It's 20 milligrams, right, right. Yep. So you never gonna get there that way. It could be it could be quite expensive. I mean, CBD is still 50 milligrams, you can take that every day, it's going to add up, but it's still promising. Well, it's pointing toward a pathway. And maybe some other tweaks in the cannabinoids. Like, maybe you got a heptyl sidechain on the CBD maybe maybe be more potent? I don't know. But it's definitely showing that it can, it can restrict the growth of the virus in those settings. They also then looked at some studies in mice. In then they looked at patients that were on sort of epidemiologists, Epidemiology, how many patients out in the world are on CBD? And do they have a different COVID positivity rate, and they saw some signal there, which I thought was, which I thought was very impressive. But there's another angle of this, I hope, when you have mine, to bring up to him is that there's the viral stage of the disease is one problem. And you definitely want to try and fix that. But after it's been in there, it's cleaning up the mess of the virus like there's this long stage of like cytokine storms that occur, not when the virus is there. But after
Nick Jikomes 1:06:32
joining, what is that what is a cytokine storm,
Kevin McKernan 1:06:35
it's your immune system goes crazy and starts attacking sometimes itself because it's been revved up to fight this virus. And the virus has got all these broken parts floating around. Now that's only replicating and it's there's a lot of high interleukin six, there's just an over aggressive immune response to try and clean up the mess. And that has proven to be more deadly than the virus itself. Right. So there are a lot of these early therapies. They're not just about using like ivermectin or monoclonal antibodies, they combine them with dexamethasone and other steroids to try and tune down the immune system. So we don't overreact to the cleanup process. And there's some very interesting work from a gentleman by the name of Muhammad as the last name of the paper where they're looking at using THC to reduce the impact of what's known as staphylococcus endotoxin V. All right, this is a very super antigenic peptide that comes out of the staff genome, however, it's got 20 amino acids that it shares in the SARS genome, which is bizarre, it's in there. But that's known as a gentleman named Chang at all published in PNAS, the fact that that sad domain in SARS is the reason why we get cytokine storms. And so right there, right adjacent to the the fear and cleavage site that's so unique to SARS. So what is that? And how do we deal with that? And will the cannabinoids actually help deal with the cleanup process as well, I think that's such a great job looking at the concentrations needed to reduce viral load. But we should also be thinking about can CBD example is playing a role in comorbidities, right. But if you look at a lot of the comorbidities, COPD and obesity and chronic pain, and a lot of these things are inflammatory diseases that make you more prone to SARS. So I think there's a whole role of cannabinoids in that front. And then secondarily to just reducing the viral load. Maybe ivermectin and other tools are better at reducing the viral load. When you look at some of the there's certainly more clinical trials, right. But the cleanup stage is where I think the cannabinoids are probably going to have the bigger impact because they're known to modulate cytokine storms. And I think there's actually direct evidence from this gentleman from Mohammed showing that THC in particular was very good at driving down the cytokine storms that were generated from SC s CB is the the acronym for staphylococcus Enterotoxin. B, this is the the super antigenic peptide that we know is in the SARS virus that's also in staff that creates these these types of toxic shock syndrome like events that we get. So I think there's there's more than just the viral load thing going on that we have to pay attention to. There's treating the comorbidities and perhaps treating the the hyperinflation that can occur after the viral infection that that may be playing a role.
Nick Jikomes 1:09:16
Interesting. Yeah. So so yeah. So when people get really, really sick with COVID 19, they typically get these things called cytokine storms. And it's just this sort of hyper overreaction of the immune system. And in some sense, it's your own immune system that's killing you because it's overactive.
Kevin McKernan 1:09:31
Yeah, D in that that's what that's what drives a lot of the buildup of fluid in the lungs and the virus is gone by then. This is why I think some of these remdesivir trials aren't working so well. They don't give people REM desert until they're in the hospital. And it's too late. The virus has gone by then you treating and that's something that stops replication of the virus, right. It's not at that stage they need they need dexamethasone and other things to tamp down the immune system.
Nick Jikomes 1:09:56
So on the subject of COVID, you know, we've now We've now got these new type of vaccines that everyone should be at least somewhat familiar with, at this point, the Pfizer and the Maderna. Vaccines are the are the mRNA vaccines, meaning they literally contain inside of them small stretches of mRNA, which in this case, encode the spike protein of the virus. And the idea is, you put these this mRNA into your body, your cells suck it up, they then make the spike protein and only the spike protein from the virus, you don't get the full virus inside of you. But that spike protein can then be presented to your immune system. And that's what gives you this immunological reaction. Now, my question for you, Kevin, on on the genomic side of this is, if you take out the mRNA from one of the mRNA vaccines that encodes a spike protein, and you put it side by side with the equivalent mRNA from the virus, are they exactly exactly the same? Or are there differences between them,
Kevin McKernan 1:10:51
they're very, very different. So first, let me just qualify this by I'm probably more vaccinated than anybody because all of my lab work and dealing with these bugs you see behind me. So I've had to get I've had to get a pin cushion for vaccines. Because of my days of Whitehead, we handled a lot of viruses and had to get back to that prevalence. So I'm not anti Vax. When I come out and say this, there are very material differences between the vaccine RNA and the actual viruses by protein. And we shouldn't conflate all vaccines here, the two that have messenger RNAs that are delivering them with lipid nanoparticles are kind of unique in this regard, and that in order for them to do that delivery, both Maderna and Pfizer decided to swap out the urine genes in the mRNA, with a modified useless site known as n one methyl site, pseudo URI. That's a different base. That's the basis a little sloppier. It has much higher melting temperatures than your dean. So if you make an oligo, with yearnings all swapped out, even just four of them inside of a 25 minutes, it will change the TM markedly like more so than swapping from for remedies to periods. So it has a huge impact on the melting temperature.
Nick Jikomes 1:11:58
So is that does that basically mean that this is affecting the stability of the mRNA molecule?
Kevin McKernan 1:12:04
This is the reason it was put in is that the folks at Penn realized if you the rnases can't cut this, right. So rnases Oftentimes target the yourselves, you clean the RNA and it clears the RNA out of the body so that you get really a femoral expression of the RNA. And that can be important for a lot of cell biology and that they the cell is anticipating these RNAs to get expressed at a certain level and also decay at the same level. Because timing on making these things is important for for cell circuitry. So when you when you then introduce an RNA that doesn't decay at all sits around for long period of time and hyper expresses inside of cells, it can throw off other other types of cell biology. But their goal is this is early on, they didn't know their biggest concern was the RNA would get eaten going in and they just have no effect. So they decorated every change every single uracil out with with this pseudo URI, which means it doesn't decay as quickly. However, it also means that the translation fidelity is affected, because now you have a different base in there. And the codons were relying on your zils to teach them what amino acids they should be putting into the ribosomes. And they've shown when you swap that out with pseudo Yardi, the translation fidelity goes to hell. And so we don't really know that it's making the exact spike protein that the virus has, because the coding system has changed. I mean, has anyone looked? That is the biggest problem, I think with these vaccines is there's very little documentation of what peptides are being made in vitro. The EMA has a document out asking Pfizer to clarify why they have smears on their western blots of these things. So they do an in vitro transcription and translation reaction, which is a model.
Nick Jikomes 1:13:40
So so when you say smears on the western blots, what that what that could mean is that instead of producing just the spike protein as it exists in the virus, it's actually producing a range of sort of variant variations, like a library
Kevin McKernan 1:13:55
for perhaps truncate right. I mean, one, one other aspect they did, which perhaps Hindsight is giving us some knowledge on this that they code on optimize these as well, which means when you're coding for an mRNA, there's multiple different codons and code for amino acids and different organisms use these in different patterns almost like dialects in a language, right? So humans tend to use a particular codon for arginine which is CG CG, CG G is the most common codon usage for arginine. The virus uses a different one, to code for RGB. And so they they change the the we have a pre printed on this, where if you just take the GC content of these two of the mRNAs that are in the vaccines, and compared to the fires, they ship radically because they code on optimize them well, when you code on optimize things. The ribosomes move at different rates. And when they arrived, because they're searching to find the anti codon tRNA and if you happen to use a really rare one in a human cell, then it has to go find that rare codons put into the ribosome and read it. And so they move these away from the the codons that the viruses had evolved to use, which I think maybe that was a mistake, right that the viruses chose, though that those codon frequencies for a reason. And maybe that's part of its functionality is that it used rare codons so that when humans make it doesn't make too much of it. They've been hyper charged this thing by using all humanized codons so that it should make even more spike protein. But more sparing protein may not be good. Right? Spike protein is what is believed to be the most toxic protein in the in the virus and making a lot of it while that might drive and accelerate immune response, it might also drive a lot of pathology, that's why you don't see the virus. So when they did this codon optimization, the GC content of the RNA is changed radically. When you do that there are the RNA doesn't fold the way it's anticipated. And you form all of these quadruplex G units inside of the RNA. And that affects all types of other cell circuitry. But the ribosomes also don't pause the way they normally pause. And they've shown that if the rhizome rate of translation changes, the folding of the protein on the back end of this doesn't happen the same way. So while you might have the same amino acid sequence, given the you didn't have the fidelity issue was to do your game soon, we assume the singularity doesn't make any mistakes, and you get the same amino acid sequence, it can still be different, because the folding is differently. And there's a great paper on this about ribosomal pausing, they actually can see this happening right in the fear and cleavage site where they change the amino acid sequences to be very human humanized, that the the folding between the s one and S two changes because there's ribosomal pausing that happens because of that codon change. So it's very complicated biology that we often, you know, the whiteboard of this where there's the codon table on mRNA, and the stop codon and the start codon. And it makes us capture everything really doesn't capture the complexity that goes on in the biology. So well, you know, my biggest concern on them is we don't have good evidence of what these things make, I cannot find, I hope someone can correct me on this, because I've been looking through the literature, there's so much COVID literature, I'm probably missing it. But the fact that the EMA couldn't find it either. And Pfizer hasn't responded to these requests makes me think the data is not out there. Which is when you put the student in vitro transcription and translation, or in this case, just in vitro translation reaction, you get a smear on a gel, not a band, you should get a single band. In fact, the DNA vaccines give you a single band that the DNA vaccines don't have superiority in them. So we can see the the like the adenovirus vaccines, those things they've demonstrated, they make a single band. And sometimes these bands are not one band, because they're the glycosylated. And we understand that, but what we're seeing on the Pfizer one, at least in this EMA document is there's a smear, and they don't know what it is. Which means we're injecting people with these things really, really what we're injecting was with a pro drug, where the active drug has not been fully characterized.
Nick Jikomes 1:17:50
I see. So So to summarize, so far RNA, we're talking about mRNA. And the four letters that go with RNA that are used in RNA molecules are a, U, G, and C. So the mRNA, that encodes the spike protein is found and in the virus itself, uses Aug and c. Now, the mRNA, vaccines for Pfizer, and Maderna, use a G and C, but a different version of the U letter. And the reason they had to do that was to stabilize the mRNA molecule, so didn't get chopped up so quickly, that there was no immune reaction, because they're trying to make a vaccine that does that. The consequence of that, that we haven't fully characterized yet, because apparently, no one has done the work to fully characterize it is that this can actually allow that mRNA that contains that that other kind of nucleotide to make a different form of the protein than what we think it's making meaning it's going to make the protein we think it's making but potentially also make other variations of it. And that's why you see things like these smears in these western blots.
Kevin McKernan 1:18:53
Yeah. And there's some evidence beginning to build up another papers I think, Bansal at all has has a figure in their paper that that tries to characterize some of these cells and they do see a couple more bands. That is a great paper, which shows that some of these spike proteins are circulating exosomes in your bloodstream for months later, the vaccinated, the Bruce Patterson's labs and a lot of work looking at long COVID. And they're starting to see in the vaccinated there's a mutated form of s one that other papers are dying to see the papers about to come out in the next couple of weeks, but in a preprint form. But they've also noted that, contrary to being infected with the virus, there's certainly long COVID going on there. They're seeing some kind of long COVID like symptoms and people that are vaccinated and there's different s one protein floating around. We don't yet know if this is because of the mRNA or this is like just like constellation is different or waiting dynasty with the papers to say about it. But there's another paper mentioned from Jang and in our preprint that looks at expressing some of these things and getting some some very smeary bands. And then there's the EMA document as well, where the regulator's went in and we're asking for Why are there smears is because the RNA also was fragmented because when they synthesize these RNA is not always full length, right? So it could the smear could be consequence of there being non full length, messenger RNA floating around. So you get a truncated peptide. It could also be a function of the fact that, you'll notice, if you look at these sequences, there's not just one stop codon, they put in two and three. And I think it's because they knew that when you when you replace the stop codons with sudo, Herodion, something that's quite common in the literature, is that it can create frameshifts over the stop codons. And then when it gets there, it doesn't block codon because it's pseudo urethane, and just Bumbles and frameshifts, and starts making something else after it. So there could be some elongated peptides, that in the back end of the Pfizer vaccine, there's actually some human peptide sequence in there, it could frameshift over those codons, which is a big if it could maybe make some of that, and then you could get some autoimmunity going on. But the thing is, this is all hypothetical and some of the stuff and we really need peptide sequencing on what these things make it humans. And some of that work is starting to get done by Bruce Patterson's Lab, which is, which should be really exciting to see.
Nick Jikomes 1:21:07
Interesting. So the time we have left, let's switch gears a little bit. I know that medicinal genomics you guys have also been doing Salafi mushroom or magic mushroom genome sequencing. So what was the impetus for that? And what does that genome look like?
Kevin McKernan 1:21:23
So that's a good question that kind of resurfaced. So many years ago, my father was struggling with with cancer, and I started reading a lot of these JHU cancer depression papers. Not necessarily for him, but I think the rest of the family that was kind of dealing with this, but I sequenced the genome way back then and tried to put it public anonymously, which is a train wreck. I had a really hard time doing that. It's really hard to put sequence data public anonymously. But anyway, that that got us a little bit interested in all the work they showed on how profound those compounds are for depression. Since then, some interesting papers have come out of doing in New England Journal medicine showing comparing psilocybin against other traditional SSRIs and it has a remarkable safety profile and performs just as well, if not better. So in this last few years, when COVID showed up, I realized there was one there's certainly a lot of depression that's going on, but there was some papers showing that blue box to me, which is an SSRI was having remarkable success and COVID like now I think they just finished a 9000 person trial and WashU is now even recommending it. But phlebotomy isn't just a serotonin stars or serotonin reuptake inhibitor it also hits the signal one Alpha receptor, which is something known in DMT. E, and DMT and psilocybin are really closely related. There's just like a methyl group that's that's different between few of these molecules, right, they just they change the methyl group between like psilocybin solution originated and that is
Nick Jikomes 1:22:50
I'm glad I didn't expect you to bring this up. But that was that was also interesting to me to the Vox mean being a sigma one receptor agonist I believe because that is a thing that differentiates it from other SSRIs I believe right
Kevin McKernan 1:23:02
is it is yes and that is you know, I'm not up to speed on all of the different try cyclists that are out there right now but that was something I picked up on reading some of these these papers as well that that's that's that's that might be more related to what's going on in SARS and serotonin but serotonin is also very involved in platelets, right so the platelets makes make their own serotonin and and that's that's something I didn't really, really understand respect until recently, the gut microbiome is credibly involved less serotonin receptors in the gut. So I don't fully understand all the biochemistry behind why fluvoxamine is doing what it's doing. However, I do know that psilocybin is likely a mimetic of it, and maybe even a safer one. In that there's there's the if you look at the the metabolism of these drugs that psilocybin is, there aren't toxic doses of psilocybin, you might have a bad experience, but you're not going to toxically overdose and so that's really promising. So I don't know I just got I got a reinvigorated to go look at this again, let's just go and sequence profile all of these. These genomes I also the first time we did this many years ago, I wasn't I didn't totally appreciate the the entourage effect that could exist in Russia, like now we know that their core beliefs and there are all of these other MLAs is monoamine oxidase inhibitors and some of these mushrooms that might change the way that metabolize it. tryptamines we also know a lot more about origination and nor bio cysteine and it seems like there's another a whole nother story very analogous to cannabis where it makes maybe six of these different psychoactive tryptamines and then a host of these other, you know, p 450, modifiers or monoamine oxidase modifiers that change the way that these these compounds get metabolized in your system. up there's a tremendous amount of therapeutic potential on this thing. And the genome is much easier to sequence in cannabis. Right. So it's only 46 million bases. It's not nearly as repetitive. So that was a story. We had high molecular weight DNA from the sample here and we couldn't get during the pandemic, we couldn't get access to PacBio sequencing because we outsource it all. And it was just backlog for sequencing suspect SARS. And so I called up Seth saying, if he asked me, Seth Crawford, does he have any extra space on one of his next runs, because he's doing a ton of sequencing on cannabis. And he said, Yes, and that over, and he blew it through his Hi Fi pack biosystem. And like a month later, we had 32 kanji beautiful reference genome. And since then, we've been that was done on the penis envy genome, try writing a scientific paper with that name in it.
It gets some chuckles. And so we then decided, we want to try and get on the perfect chromosome stretches. So we then put a what's known as a Heisey. Map across this, this is a technique that's really cool. It takes DNA from an organism, we had to do this in spores, which is really tricky, because we can't handle tissue here. The spores are legal for taxonomy purposes. And we think DNA sequencing falls under taxonomy. So what we took is we took spores, and you treat them with like, crosslinking agent that crosslinks proteins to one another. So this takes the proteins on DNA and glues into one another, which means DNA that's in close proximity is most likely to glue to itself better. Why do we want to do this? Well, when we finished sequencing the genome with Seth, we had 32 pieces of the genome, but we didn't know which ones were in the crop, which ones belong to which chromosomes. And so what you can do is do this cross linking trick that will tell you which one of those pieces are in proximity to one another in the genome on the chromosomes. And these Heisey maps do a wonderful job at binning the context of the genome assembly into context, or your chromosome one through 13. So we know now we know we've got 13 chromosomes in the genome, and a mitochondria, it's about 46 million bases long. And we've got a beautiful reference grade genome for for the penis, semi genome, and then we then we then went out and sequence 81 other genomes from B plus to Golden teacher to you know, you name it, and then mapped all of those reads back to that, that original reference genome to look for variants. So we now have a good catalogue of all the variants of the most common cubensis lines that are out there, and what variants exist in those genes. So you'll probably see a month from now, I can a PDF for that comes out and does the same thing with philosophy as it does with with cannabis. And that will, we'll have coverage maps over psi m psi k psi d, psi H, these are the four genes that are kind of analogous to CBD, THC and CBD synthase in the in the in the cubensis genome. So we can look for variants and those that might correlate with chemo types. We're kind of like five years behind cannabis on this front, because we don't have good chemo typing right now, there's only a few labs in the world that are that, that do this type of chemo chemo type thing. And sadly, we haven't sequenced any of the samples that have been chemo type. So if there's anyone out there that does that call us we want to sequence any sample of chemo type. So we can start from these correlations between genotype and phenotype.
Nick Jikomes 1:28:04
I, I think I know someone working on that. So I'll see if I can I haven't talked to him in a while. So I'll follow up.
Kevin McKernan 1:28:09
That'd be fantastic. Yeah, it's sorely needed in that field. But we've got a great map of all the variation out there. We just can't express any of this tissue. So we have to rely on other labs that do that and correlate the results.
Nick Jikomes 1:28:23
So the other thing I wanted to ask you about is just the general, I just wanted to ask you about blockchain technology. So I know that you guys use it at medicinal genomics, I believe. So can you describe how you guys use it there? And actually, before that, just just give people a quick blockchain one on one?
Kevin McKernan 1:28:40
Oh, yeah. So blockchains are horrible databases that are distributed all over the world. They're very information expensive. So you don't put stuff in blockchains unless you want it to never change, and they don't hold a lot of space. So we would never put our entire genomes in blockchains. Because no one wants to replicate our genetic sequences on millions of computers around the world and synchronize them every every 10 minutes. What blockchains are great at doing are storing perhaps fingerprints of larger datasets, hashes of them as what they call them, great for the cannabis field, another hash. But a hash is just a fingerprint of a larger file. And it's a unique fingerprint being you change one letter in the original file and the hash changes. So what Bitcoin is really doing is it's storing hashes of transactions. So that when you when you it's a big ledger, that everyone replicates the ledger all around the globe. So when I send money to one person, all computers update that I sent money, that person and there's no disagreement over me sending that money to somebody else. And they synchronize on a certain time scale, but you know, bitcoin does it every 10 minutes. Other networks do it every two minutes. You can pick a blockchain that does it faster or slower based on your needs. There's there's even layers built on top of Bitcoin, like the Lightning Network that can do Instant transactions. But what we're using it for is a way to timestamp when data was generated. For people that's in a database that we can't have, we can't control, right? We don't want to have a company who is doing sequencing that declares when a certain sample was sequenced and what it's related to. And if someone can call me up on the side and give me 1000 bucks to go change the timestamp, or rearrange the data, right, where we go out of business, and then all the information leaves, right, the thing about blockchains is, is your financial motivation for those things to persist in time, that will probably help last a company's lifespan. It's been around for 10 years now, we've been around for less than that. But there is a monetary incentive to maintain that blockchain all over the world. And it's in a way where you can't just knock out one computer and have it go away. You can't censor the stuff. It's immutable. So mutable records are very valuable for discerning lineages of things, and timestamps of things. And whether that be DNA sequencing, or patent information or dates of invention, you can use it for a variety these things, some people use it for FFTs, right. So what we're doing is we're taking people's variant table, we sequence for them, we're creating a hash of that VCF file. And we're spending that into the opportune, which is a field in the Bitcoin address is not the address, but a field in the block structure that you can put like metadata and we stuffing into there. So it then gets woven into a blockchain transaction and can never be deleted in time. So now we can give someone their DNA sequencing file and say, you have your information, you can run Sha 256 on that file, it will create a fingerprint that exists at a timestamp in a blockchain that no one in the world never edit. So you can prove that your file existed in a court of law by just showing that your sequence file creates a hash that has a timestamp in Bitcoin. So it's a it's kind of a digital notarization system, when you can't go to a bank to get your plant notarize. Right. It allows you to kind of bypass the whole, you know, formality of getting someone to put a weird paper print on something, you can just say, you can prove in the court of law that no, my plant had to exist this time, because there's a hash that exists in Bitcoin that matches my sequencing file, and nothing else in the world could have generated that and no one could have ever mutated it.
Nick Jikomes 1:32:05
So blockchains are they're not good at storing extremely large volumes of data. But they're good at storing small volumes in an extremely secure and immutable way. So if you just want to track when something happened, and you want to track those timestamps in a way that can't be tampered with a blockchain is your best bet.
Kevin McKernan 1:32:24
Exactly. Yeah, you summed it up much more elegantly.
Nick Jikomes 1:32:29
And so I guess the way that it would work for Bitcoin is, you're actually not storing an incredible amount of information in the Bitcoin Blockchain, all you really need to record is like this much Bitcoin from this address to that address. And it's really a small amount, yes, very
Kevin McKernan 1:32:41
small amount. And we, you know, there are times in our we've been doing this for five years, and there were times when bitcoins transaction cost was like 50 bucks a transaction. And so we, we pivoted to use other blockchains. And at the same block structure, like Dash has the same block structure, so it has an operator and we can go into dash, we can also go into bitcoin cash, and we can go into the theory of those those for blockchains. We've got API's that can, so people can point, it's whatever chain that they feel is their favorite chain. But the the important point there is we're just putting a fingerprint of the file into the chain. And to do that, you need a very small amount of very small transaction fee, to stuff something into the the opportunity is the minimum amount, that smallest amount of what you do. And then, you know, there's usually though the way the Bitcoin network operates is all of these suggested transactions go into a memory pool that all the computers look at that are mining, and they sweep from that pool to make the next block, they're in a competition to make that as fast as they can the first person to make it gets the reward. So there's a bit of a fee market, if you will pay by fee market that you can, you can attach a higher fee to your transaction. And the miners are going to, they're basically going to select what goes into the blocks based on the highest fee structure because they make more money that way. So the fees can vary on this based on market demand, which is why we have alternative chains, people want to put them into the chains.
Nick Jikomes 1:34:01
Interesting. So Kevin, what what are you most excited about? Not necessarily in your own personal work, but just in the world of genomic generally what's going on right now that really excites you?
Kevin McKernan 1:34:13
Well, I love the these chromatin capture methods are awesome for cleaning up genomes. The long read tools we have with PacBio now are exceptional in that they assemble plant genomes like practically a laptops emulations a lap of the fungal genomes, we did a laptops, the cannabis ones, we don't need these massive compute centers now that we have these really high quality research pack bio. There's portable sequencers coming out from Oxford. They're they're not as accurate as Packfile. But they're exciting and that they're very small. And we could probably find some really new applications for that future. And then I'm really excited about a lot of the microbial work we're doing and that we're now learning we have a huge database now of all the my microbiome is we found a cannabis and stuff you can see behind me. We're constantly sequencing these things, putting them on different plating mediums to see which organisms grow where and which ones Don't and we're learning a lot we're learning like what pathogens do don't grow on plates that we can find with DNA. And some of these are pathogens. You know, one that's in our most recent study is Pantone Pantone Penta agglomerations is actually published as a plant promoting path, micro by cannabis. There's also some evidence in the literature that if you get this inside a human cuts, like it can give you sepsis and there's there's problems with it. That's an organism that's like we find in high prevalence, a lot of our microbiome studies, and we now know, it doesn't grow it unless you grow it down to 30. C, at 36. You can't you can't get this thing to grow. So we know that there's, there's microbes out there that are pathogen screens are missing that you can only get with PCR. And that's interesting to us. We've seen this on like for other organisms as well, where they're, they're just not growing unless you grow them at lower temperatures. But they do find themselves in clinical cases at the CDC, like there's, there's, there's this pathogenic event so we're beginning to understand like which pathogens we want on the plant that help for promoting the growth and which ones are benign to humans and which ones are actually pathogenic to humans and, and how to separate those with with the appropriate tools out there. So I think that's where we're gonna find a lot of both in the philosophy space. And I think in the in the cannabis space, like understanding the microbiome, and the fungal, like mycelium, they're running these things, I think is incredibly important. I mean, we've, we're seeing some evidence now in the literature that even like these, these hops, hop, lean bioroid. In other crops, some of these by roids actually travel to fungal mycelium networks. Alright, so we have seen with at least Hopland bioroid, when we go and sample that the plants that are affected with that, it's in the stems, it's in the leaves, it's in the roots, right? It's in the same copy. Like that's, that's really bizarre to us that we find universal expression of this thing and every tissue doesn't make us think that maybe it's at least we've heard from some growers that have plants in one part of the grow in they're not in any mechanical contact with another plant. They can see the other plant get it. So it's not known to be aerosolize. So what the hell's is making these things communicate this virus if they've been really careful cleaning all their scissors, everything else? I don't know, I think there's a lot we need to learn about how the plants interact with the rhizosphere and in the mycelium, and it's probably going to play a role in virus transmission, even some of these other bacterial transmissions that are going on. But that's an area that I think is just, it's wide open look very little is known about it. And it seems to play a really critical role in how we regulate cannabis. Like what are we doing from a microbial testing standpoint, we shouldn't be failing people if they have like beneficial microbe use, because that's going to push them into pesticides, which are a whole nother, you know, probably train wreck going on beneficial microbes. We shouldn't have this world where we're afraid of every single micro organism out there. Like I think we're I think we're absolutely hypochondriacs when it comes to the SARS virus because the vast majority of people it's not symptomatic, right. And maybe there's an aspect of human biology. We haven't fully appreciated here that we need frequent introductions of certain viruses to like boost up our mitochondria or something. There's probably a role for these things on mutualism that we're not fully understanding because we're panicked. But I think we find the same in these cannabis micro biomes that were hypertensive to maybe Aspergillus, but there's probably some species of Aspergillus that are beneficial to the plant that we shouldn't be penalized in people for. The same is true with bacillus, right. This is used as a common microbe that is beneficial to plant growth, and probably innocuous to human health. So I think teasing apart that is what we're really excited about here. This was genomics because it kind of fits with all of our are motives here. We're making PCR tests that help diagnose these things. And we want to find the genetics involved in the plants interact best with them. So, so lots to unravel there.
Nick Jikomes 1:38:51
What about What about something like CRISPR? How far away are we from having you know, crisper cannabis called I think some
Kevin McKernan 1:38:58
people have done this at least if you read the patent literature now. That's that's the patent literature is in peer review, you know, you can try and patent anti gravity devices, right, but, but the methods in those things look like it could have pulled it off. Like we have the genome resources now and the references for people to do it. I guess the question is, you know, that's going to be regulated differently because there's different opinions on GMO in different jurisdictions, right, like Europe has a different definition that we have here. I've personally been more excited about like the triple A to work because that had immediate benefit. Like one the triploid work. Um, you can argue it's a GMO event, but it's more classical. We've done it in a lot of other agricultural plants to get sterilized plants. But this is something that looks like it's boosting yield. It looks like it's making seedless plants and it's ready today, right? The CRISPR stuff, I feel there's an excitement for people to do it because they know they'll get something as patent. Right. You've man modified it now, you know, it's patent eligible, and it's somewhat unique for a few people that have before I think is going to be some resistance to it. And from the consumer front, I think it'll be less of that resistance to triplets. Because everyone's had seedless watermelon before, right. So I don't know, I'm a little bit more attracted to some of the breeding techniques that don't trigger people's, whether they're irrational or not. They're they're like, there's people who don't want GMO, and you got to respect that consumer base. And I think there'll be a portion of those people who don't want GMO that will accept a classical AAA plan, because they've been in the food system for for decades now. And they don't see them as carrying, you know, foreign DNA per se. Now, you can also use CRISPR, to delete things. And that brings up a whole other top right. So it's not just about adding foreign DNA in or modifying the DNA in a way that's already seen naturally elsewhere. Like you could go in and make some of these modifications to THC. synthase, right. And they could emulate things that are in the wild, but very rare. People may have deep philosophical opinion over that. Or you could delete a gene where you're not adding in anything new, you're taking something away. I think some people may have less concern over that than like, I'm putting in a THC synthase that also has a GFP on it, right? It makes all the males glow or something, right. I think those things freak people out more when you bring in like, you know, a jellyfish gene as a marker to to help facilitate your selection of the cells that are actually modified, right? Those things, I think, scare people more than just Alright, I'm gonna knock out a gene, that's a problem. I think there's a lot to be done there, we have the tools to do it. I just think the markets gonna need a lot of educating on the different flavors of modification that go on, and whether they're philosophically aligned with them.
Nick Jikomes 1:41:41
Well, Kevin, thank you for your time. We covered a lot of topics. I think there's a wealth of information buried in this episode. So if you have any final thoughts you want to leave people with, go for it. Otherwise, thank you for your time.
Kevin McKernan 1:41:54
Oh, yeah, absolutely. There's a combat conference that we run in Pasadena, it looks like it's going to happen this year. It's been postponed for COVID for too long. So it's happening in Pasadena. Check our website, they'll have dates and information on it. It's got forgotten which dangerous things may. Now we punted it so many times. But this talks about a lot of we get speakers from around the world that discuss all of these topics. And there's people well more versatile than I am particularly on the medical front on how to deploy cannabinoids and even the analytical chemistry front. So that that gets that gets a lot of conversations like this going that are really fruitful. So if you're interested in getting back to conferences, we're having one and it'll be Southern California this spring.
Nick Jikomes 1:42:34
Excellent. Kevin mckernon. Thank you. All right. Thank you. Take care.