Transcript

Genetics - Why Most Of What I'm Going To Tell You Is Wrong: Dr. Thomas Merritt at TEDxLaurentianU

[Music] I'm a geneticist. I study how genetic complexity is transformed in the biological diversity. How the information in a cell is transformed into the biology of the cell. How do the slight differences in genetic code change into or transform into the differences between individuals? Look around this room. We're an incredibly diverse group. And a lot of that diversity comes from genetics. In biology, genes are the packets of information, the words. The letters are DNA. And different forms of a gene are the alals. Think brown eyes, blue eyes. Different forms of the same gene, different alals. Genes are strong along chromosomes. These strands of DN D D D D D D D D D D D D D D D D D D D D D that we find in the cell. The entire set of chromosomes in an organism is its genome. What I study as a geneticist is how that genome functions. In 2001, we published the first draft sequence of the human genome. And one of the first things that we found was that there were too few genes. A lot too few genes. Based on the complexity of us, we had estimated there'd be about a 100,000 of these packets of information. About a 100,000 genes. And we found 25,000 genes. We were off by a factor of four. We knew we wouldn't get it exactly right, but we didn't think we'd be that wrong. So, how were we wrong? How was it that we had underestimated? Try it again. The overestimated underestimated we guessed on too many genes. Why were we thinking 100,000 and why did we find 25,000? Well, it ends up if I were giving this lecture five years ago, I'd have a different answer for you. And if I give this lecture in five years from now, I'll have a different different answer for you again. So much of what I'm going to tell you tonight is in fact wrong. But this is how science works. We take the facts that we have, we put them together, we try to understand the world around us, but we have to appreciate that much of what we're guessing at is in fact wrong. It's a misconception. It's a simplification. Sometimes we're just wrong. When we estimated that there were 100,000 genes in the human genome, we were wrong. But how we were wrong tells us quite a bit about how those genes actually work. The exciting thing for me as a geneticist is just how much we're realizing how wrong we are. But it's not just how wrong we are. It's how are we wrong that's driving science. In what way are we wrong? How have we been misinterpreting the world around us? Once we can start to answer that, then we've got a better chance of figuring out how could we possibly be right. So, what was wrong with our estimate in the number of genes? The first thing that we was that we had incorrectly estimated how complicated any single gene was. And what we had naively thought was that there were a lot about 100 thousand really simple genes and they worked together to give us the complexity of the biological world around us. It ends up that there were a lot fewer a lot more complicated genes that were interacting to give us biological complexity. So how are the genes more complicated? One of the things we were wrong in was the fraction of the genome. Remember, all the genetic material in the cell that's actually dedicated to genes. It's 2%. 2% of your genome are those packets of information that make up biological diversity. 10 times that 20% for those of you like me or they're not good at math is involved with regulating where those genes are active. Now all cells essentially have the same genetic material has the same complement of DNA. They have the same genes. What differs cell to cell to cell? What makes a skin cell a skin cell and not a muscle cell or a nerve cell is what genes are active in that cell and it's the regulation of those genes that drives biological complexity. And what we were missing was how complicated that regulation is. One of the first things we had wrong was we generally had this vague idea, maybe not so vague, that we had genes on this chromosome and upstream of that gene, there are little switches, little regulatory elements and proteins bind to it and they turn a gene on and off. Well, it ends up that that's basically right, but it's a simplification. And like all simplifications, it's wrong. And it's wrong often in really interesting ways. The first thing was that those switches are more complicated than we thought. So it's not just a switch, it's a series of switches and those switches interact together to more finely modulate those genes that we'd expected. In addition, what we're finding now is there are a lot of novel mechanisms that are outside of our traditional models for gene regulation that are regulating those genes. These novel mechanisms we refer to as epigenetics. epi above beyond around traditional genetics and in the last few years epigenetics has been receiving a lot of really welldeserved attention in its role in driving biological complexity. Now, as I say this, it's important to realize that what's changed is our understanding, not the mechanisms. Genes are regulated the way they've always been regulated. It's just that we're better understanding how they're regulated. We'd simplified the model. We were wrong. We're learning a little bit more about it. Maybe we've got a chance of being a little bit more right. In my lab, we study gene regulation in flies. Fruit flies. Drosophila melanagaster. You may have played with them in high school. You certainly have had them fly around your fruit bowl in the summer. No, I do not know how to get them out of your house. I can't even get them out of my office. One of the things that we study in fruit flies is how the architecture of the genome, the shape of the genome drives gene regulation. This work has pushed us beyond the bounds of traditional genetics and forced us to change the models of how we think biology works. The first way it's done this is it's made us appreciate that chromosomes have three-dimensional shape. They have architecture. They have volume. Traditionally, we've thought of chromosomes as being these passive clouds of information, these strands of DNA in the cell in the nucleus. We now know that these chromosomes are dynamic. They have shape. That shape is regulated. They have location. Chromosomes f function more like organs than they do like clouds of passive something or other subcellular organs. We call them organels and they're organels that interact with their environment. So what do they interact with in flies? One of the ways that chromosomes interact is with each other with other chromosomes. Flies are diploid. They have two copies of every chromosome just like humans are diploid. We have 23 copy, 23 chromosomes, two copies of each, one from each parent. Flies are essentially the same except they have four chromosomes, but they have two copies, one from each parent. We call those homologous chromosomes, the copies. In flies, the homologous chromosomes pair and they pair a lot. They're essentially always together. And that pairing is a part of gene regulation. We recognized this in the 1950s and it was described as a phenomena called transvection and it's a dependence of one chromosome on the other. 60 years later we're beginning to understand the role of this transvection this pairing dependent regulation in biology. The different chromosomes are not a dependent. The experiment that we did in our lab, my lab, we had created, we genetically engineered a set of flies where we deleted a piece of the chromosome and we turned off the gene and those work fine. And we do this because we can study what why a fly needs or why we need a gene by turning it off. We do it in flies because we can genetically engineer flies. The surprising thing was when we paired these engineered chromosomes with a normal chromosome, the regulation went haywire. and it went haywire in a pairing dependent fashion. The chromosomes had to be able to line up. This lack of independence may not seem like a big deal, but to a geneticist, this is a really big deal. This doesn't make sense in our traditional models of how genetics work. And it's forced us to rewrite the way we think about gene regulation. The next thing that we found was that this interaction, this dependence between the two chromosomes at the one gene that we were studying is modified by a whole suite of genes across the rest of the flies genome. There are about 15,000 genes in flies. We're studying one of them. The other 14,999 are modifying the expression we see in that one gene. We call this a genetic background. So our counterintuitive pairing dependent gene regulation is modified by this counterintuitive genetic background effect that comes in and exacerbates this complication. And in the work that we're just starting to publish now, we found that that first level of dependence and that second level of background is modified by a third level of the environment. If you change the environment of the fly, you change this dependence. You change the way the gene is modified is regulated. So we have an unexpected form of gene regulation wrapped in an unexpected form of gene regulation wrapped in an unexpected form of gene regulation. Those three levels are beginning to explain the complexity in genetics and in biology that we're struggling to put together. Okay. So why does this matter? At the end of the day, we're really not that interested in how flies regulate their genes, but we are interested in the global question of gene regulation. How does gene regulation lead to biological complexity? the system that we're studying flies is beginning to tell us how that gene regulation actually is accomplished. Now, ultimately the goal is often to try to understand our complexity. How is gene regulation accomplished in us in humans? What can what we're studying in flies tell us about that? Well, it ends up that we're diploid, as I said, but our pairs of chromosomes very rarely pair. Very common in flies, almost un unheard of in humans, except in some really interesting cases. One of them is in a certain fraction of renal cancers, there's a mutation that duplicates an arm of a chromosome. that duplicated chromosomeal arm pairs with the normal copy. Physically lines up, pairs in exactly the same way that we think the chromosomes and fruit flies pair. That pairing that's so important to gene regulation in flies disrupts the regulation on that chromosome in that human cancer cell. That pairing leads to misregulation of the genes on that chromosome arm. That misregulation is what's driving those cancers. Pairing dependent gene regulation underlies that tumor genesis, that cancer. How does it work? Yeah, we don't know. But what we're finding out in flies is beginning to tell us how that system is driving cancer. What we're finding in flies is beginning to tell us how we can understand genetic complexity and biological complexity. And at the end of the day, this is all based on essentially a bad day in the lab. I was working in the lab one Sunday afternoon. I genetically engineered a bunch of flies. I had some expectations on what I was going to find when I ran these assays and I wasn't getting the answers I expected to get and I couldn't figure out a way to make them work. And my postto adviser, the professor I was working with at the time came in, looked over my shoulder and said, "Huh, I don't know that might be transvection." And walked away. 10 years later, a third of my lab is dedicated to trying to understand this really counterintuitive form of pair interaction that we've demonstrated is completely dependent upon the genetic background and highly modified by environment. All because a set of almost chance observations. This is how science works. You spend the hours in the lab. You do your homework. You make the observations. You know, you never know when an almost stray observation is going to change the way you view the world. And honestly, this form of regulation has opened up an entire new set of models for us on how gene regulation works. 10 years ago, we didn't realize how complicated gene regulation is. We didn't know how wrong we were about gene regulation. Now we're closing closer to understanding how 25,000 genes can do the job that we expected a 100,000 genes to do. I have the greatest job in the world. I get to explore, discover, question all the time knowing that most of what I know is wrong. But it's not that it's wrong. It's how it's wrong that's most interesting. What are we going to What are we going to discover next? We don't know. And that's what's really exciting. Thank you. [Applause] [Music]