 Good morning. So I've had a number of questions about the exam. I mean, there's a little bit of fear in the air, frankly. So the question is, I mean, people fishing for how the test is going to go. It's going to be like the other test you saw in terms of its multiple choice. People have asked the extent to which I expect details to be known. I've made it very clear that I expect you to know what comes out of my mouth in the lectures and what's in my notes, whether you're not responsible for what's in the textbook. But then again, there are quite a few names and details in my notes. And the question comes up, well, do I need to know all the stuff in the notes? Well, frankly, in all fairness, I think it's fair that if I did put a detail in there that you'd be responsible for, because it is consistent with what I said. However, I tend to ask more conceptual questions. So I'm more concerned that you understand the basic ideas that we talk about and then maybe you can give an example of each. So don't expect, I mean, that's just the nature of my tests. I'm more concerned personally that you understand the concepts rather than memorize every detail. There are some things you'll be expected to mean. There's some easy points. I'll just tell you that right now. If you memorize the geological time scale, which is what we'll get next time, there's an easy point or two. If you know the Lenin classification system, that's another thing you can easily memorize. There's a few more points there for you. But for the most part, it's going to be conceptual. Are there any questions? OK, so let's get on to today's lecture material. So today I'm going to talk about phylogeny. I wrote down an outline of what I want to talk about today. And this is to remind you what we'll be discussing precisely what a phylogeny is. But evolutionary biology is really founded on the idea that all creatures on Earth share a common ancestor. And we discussed some of the evidence for that. But that's the basic idea. And that, of course, natural selection find tunes organisms for their environments. But the basic idea is that all of organisms on Earth are related by a tree, a tree of life, if you will. And this figure on the right, rather, is the only figure in the origin of species. You may have already seen this, but it's basically Darwin sketching out what a hypothetical tree of life might look like for a small group of organisms. And then this was a sketch, a very famous sketch, from one of his notebooks where he sort of sketched out an evolutionary tree. And I think over here it says something like, I think, and dot, dot, dot. And who knows what he was thinking, but there's a tree. Right after the origin of species, there were a number of people that became very excited about trying to uncover what that tree of life looked like. Who is related to whom is the basic question? And so this is a tree that came out within 10 years of the publication of the origin of species by Ernst Tecla, a German biologist who is a very enthusiastic evolutionary biologist. And he made a tree of life that, if you look at the very top here, this is all in German, but this is men. And then you have a gorilla and some other. I can't really see it from here. But you have the great apes up here. We're one of the great apes. And then these are vermies as worms, and he's got other organisms on this tree. And what's really cool about this tree of life is it's kind of drawn as a tree, which I really like. You don't actually see that these days. So on to what I want to talk about. So most of this is going to be on the board. So what is our modern depiction of a phylogeny? So you'll see phylogenies drawn today in scientific papers and in textbooks as a branching diagram. So here's an example of a phylogeny. Typically, the tips of the tree are labeled with the species names. So these would be species A, species B, and species C. And the way to read this phylogeny is as a way that statements about relationships. So what this tree says is that species A and species B are each other's closest relatives. So A and B are each other's closest relatives on this tree. And the reason they're each other's closest relatives is because they share a common ancestor right here. This branching point represents a common ancestor. And on this tree, this common ancestor is more recent for A and B than any other common ancestor on this tree, the other one being right here. So it's about statements of relationship or about recency of common ancestry. One of the things that people become confused about is sometimes you'll have, for instance, if A was a human, B was a chimpanzee, and C was a gorilla. So let me rewrite these labels to reflect that. So if, for instance, this was a human, that was a chimp, and that was a gorilla, first of all, this is the tree we think best reflects the evolutionary relationships of those three species. And so our modern understanding is that humans and chimps are each other's closest relatives. So if you ask the chimp who's your closest relative, the chimp should say if it could talk, humans. And if they were up on the literature. Now some people say, but if you look at chimps and gorillas, they look much more similar to each other, at least by eye. They look more similar to me. They're hairy, right? They live in forests, they fruit, right? And humans are quite different. We walk up right, we don't have a lot of hair, and there's other features where we're presumably smarter. And so humans look a lot different than chimps and gorillas. So even though chimps and gorillas look more similar to one another, humans and chimps are each other's closest relatives, because the only thing that matters is the recency of common ancestry. Now the other thing that some people get confused is that we can rotate the branches. So if we take these two and we just flip them around and we put chimp on this side and human on this side, we haven't made any changes in the phylogeny. We're still making the same statements about the relationships. Now if I do this, if I actually detach some taxa, some tip, some labels of the tips and put gorilla here and humans over here, now that implies a different relationship, right? That says that chimps and gorillas are each other's closest relatives and that's what people thought for a long time, okay? But so this, you can do some things without changing the meaning. That is to say you can always, I can always flip, grill and chimp again. I can, that is to say I can flip these points here and get the grill on this side and chimp here. That doesn't change the meaning. So if you think about these things as being, I don't know if you ever had those connected type toys as kids that have the little wooden dowels that have lots of holes in them and you put sticks in them. Well you can imagine this being like a little wooden dowel. Actually that's kind of clever, isn't it? I actually made a wooden dowel there. Where you put a stick in here and a stick in here and the stick in here and you can sort of rotate that dowel and you don't change the meaning. But if you take a stick out and put it somewhere else, you change the meaning of the tree. Is that clear? Good. Now there's a difference, of course there's terminology. So people distinguish between cladograms and phylograms. So what is the difference between these two things? Well cladogram, the lengths of the branches have no meaning, okay? The basically the lengths of the branches are just drawn so that the tree looks nice. Now the tree of course has, how the tree is connected up does mean something. It means it gives you information about how they're related. But that's it, okay? Now phylograms, the lengths of the branches mean something. So for instance, you'll often see phylograms drawn like the following. You'll see maybe human chimp gorilla. Now what might these branches mean? Well they could mean a number of things. I'll give you two examples. The lengths of the branches might be in terms of millions of years or in terms of time. Which would be a sensible unit when you think about it because you can also think about phylogenies as representing the history of speciation events that led to the living taxa, living species. And so this tree says that there's a speciation event between humans and chimps that occurred sometime in the past. So if these branches can also mean if they're in terms of time they tell you when the speciation events occurred. So for instance I'll just give you a rough idea of five to seven million years ago would be our best estimate of when humans and chimps split. And then gorillas were not much before that. Eight, nine maybe, I don't know. I should probably check that one. But the gorillas, the speciation event between gorillas that led to gorillas and humans and chimps, this one right here, occurred before the one that led to humans and chimps but not much before. So sometimes the lengths of the branches are in terms of millions of years and the tree would be termed a filogram. More typically the lengths of the branches aren't drawn to be proportional to time but rather to the number of changes that occurred along that branch. So branches drawn proportional to time or number of changes. And of course the number of changes of watt. And so we'll talk about that but it might be the number of nucleotide substitutions that occur in a gene. So they're sometimes drawn, often these phylogenies as you'll see are based on DNA sequence information and the branches then might be in terms of the number of changes of DNA substitutions that occurred along each branch of the tree. Now that's all I wanted to say about distinguishing cladograms and filograms. I want to now turn my attention to what are called rooted and unrooted trees. Now the tree of life is rooted in time. So there's a direction to the tree of life. The speciation events occurred in a specific time in the past and time is directed, right? So we're all in agreement that the tree of life should be rooted. Here's the root, there's the root. And so if this is the root these might be the species here, let's make a fourth species, we'll make it D. This is an example of a rooted tree. Now it's interesting but it's an interesting but true fact that most of the methods that try to estimate phylogeny given you give these methods and say some DNA sequence information and they return a tree. These methods actually work on what are called unrooted trees. So here's an example of an unrooted tree. I'll label it unrooted. This is what the phylogenetic methods typically return. Now this tree can be rooted along one of five different branches, you can either root the tree here, here, here, here, or here. Now this particular tree, if you happen to root the tree along this branch right there, you will get this rooted tree out. And hopefully you can see these trees are hard to see at first but you'd have to get that knack of it after a while. But imagine grabbing this tree right here, pulling it down where I marked the X and then rotating these, breaking this branch right here so it sort of goes in a V, and you'd get this tree out. True? If you don't see it, convince yourself over time that you stare at these long enough and you'll see it eventually. It's like those pictures that have an animal or something in them. If you stare at them in a while, then all of a sudden you see it. This is the same type of thing. So anyways, just realize that these methods work on unrooted trees and that you have to use some other criterion to root the trees. I believe you guys probably heard of the outgroup criterion in lab, or you will. Basically the outgroup criterion is the following. It says, what you do is you designate one of the species on the tree as the outgroup. So let's say you're an evolutionary biologist, you become one let's say, and you're interested in the phylogeny of what? Mammals. You're interested in the phylogeny of mammals. So you go to the zoo and you ask the zookeeper, I just need to take a little blood sample from all the mammals you have. And they say, fine. And so you go to the tigers, you get some blood, you go to the giraffes, you get some blood. You go to all the mammals in the park, and you get a blood sample and you go back to the lab, you sequence a particular gene, or today you do many genes, and you wanna make your phylogeny mammals. And if you were to do that just on the mammals you sample from the zoo, your phylogenetic method, you put your data into a computer program, it would return a tree that's unrooted. And then your job is to say, well, how am I gonna root this tree? So what you do is you go back to the zoo, you say, I need something that's not a mammal. I need a species that's not a mammal. I'm gonna take a blood sample from that. So you have lots of choices. You could take a bird, you could take a crocodile, you could take a turtle. There's lots of things in the zoo that aren't mammals, but you take one of them at least. And so maybe one of these species here is, we'll put an O here for out group, but one of the species is at least one is not a mammal. Maybe you go to the San Diego Zoo and you take a Galapagos tortoise. Take a sample of the glidel of Galapagos tortoises. So you take a sample of blood from the Galapagos tortoise, you go back to the lab and you sequence that Galapagos tortoise for the very same genes you sequenced for the other mammals. And then you put that Galapagos tortoise into the same data matrix, the same data file, and you give it to the phylogenetic method. Now the phylogenetic program will return an unrooted tree again, right? But this time the out group species is in that tree as well. And the idea with the out group criterion is the branch you choose to pull down to become the root is the one that leads to the out group. So if you were to use this tree with that out group, the rooted tree you would get out would be the following. You'd have A and B, C and O, okay? That would be how you would root the tree using the out group criterion for this particular tree right there, where this is your Galapagos tortoise and A, B and C are your mammals. That's the out group criterion. That's how people turn unrooted trees into rooted trees in modern biology. That's what people do every day in universities across the world. So that's what I wanted to say about rooted and unrooted trees. So let's talk a little bit about the combinatorics of trees. Does anybody know what combinatorics means? All right, so it's the mathematical field of counting things, okay? So if you remember like factorials and permutations and combinations, you got that at some point in math in high school at least, that's combinatorics. It's just counting up how many ways things can occur. So let's add, so there's a question here, which is let's say we have four species. There they are. You say I'm interested in four species A, B, C and D and you're interested in estimating the phylogeny of A, B and C. How many possible phylogenies could explain the relationships? How many possible relationships are out there? For just four species, we can easily enumerate the four unrooted trees or three rather unrooted trees. Here they are. One, two, three. It's not very exciting yet because I haven't put the labels on, but we can have A and B on one side and C and D on the other. We have A and C on one side and B and D on the other. We have A and D on one side and B and C on the others. If you have four species and you ask one of these computer programs to estimate the phylogeny for you, these are only three possible trees that the program can return. One of these three trees is gonna best explain your data, whatever that happens to be. We'll get more specific about how the estimation procedure occurs in a bit, but this is just the number of possible results you could have had. Now, that's for four species. Let's look at what happens when you have more species. So what I'm gonna do is I'm gonna make a table. Let's see, so we're gonna make a table so we'll make it look like a table. So N and the number of trees unrooted. So this is our table. We just discovered that with four species, excuse me, there's three possible trees. With five species, you have 15 possible trees. With six, you have 105. And with seven, you have 10,395 possible trees. So notice the number is growing very quickly. You have, I see, oh no, I'm sorry, about 10,395, that can't be right. That'd be 945. Eight is 10,395. There's a trick to this. Nine is 135,135. Let's see, we make the table a little bit bigger. You go to 10 species and you have 2,027,025 possible trees. The number of possible trees is growing very, very rapidly. By the time you get 11, you have about 34 million. Somewhere around 20 species, maybe 21 species, you have a mole of trees. Remember Avogadro's number, right? And around 60 species, there's more trees, more possible trees, more possible trees than atoms in the universe. That's a big number, right? So whatever that number is, it's big. It's roughly 10 to the 60, 10 to the 70 somewhere in there. It turns out that people today are routinely working with problems that have hundreds of species in the data matrix, which means that the number of possible trees is many more than the number of possible atoms or even electrons in the known universe. So this is what's called the combinatorial explosion. The number of possible trees explodes exponentially. And this is one of the prime difficulties in estimating trees, which we're gonna turn to next, is we've got some data and that data's gonna say one or several of those trees are the best explanation for the data, the best estimate of the phylogeny, but then the question becomes a computational one. How do we screen this enormous number of possible trees to find the best trees? It turns out remarkably that you can actually do this without, obviously nobody's gonna count or go through all the possible trees. You can't enumerate in real time or even not in real time all the possible trees for 60 species, but there's computational tricks that allow you to still get very good trees out, even though that you can't enumerate all the possible trees. But this is just to give you an idea of the combinatorics of trees. And I don't know if you saw the trick, but the combinatorics goes up as the factorial, but it's not the factorial you remember, which is one times two times three times four times five. Notice it's one times three for four and then it's one times three times five for five and then it's one times three times five times seven and so forth. So you're taking every odd, you're taking the odd factorial, factorial of odd numbers. Not that you have to, don't remember that, but if you wanted to know the trick, that's the trick. Okay, so why did I wanna go to estimating phylogenies? Okay, so are there any questions on the first part of the lecture? We talked about cladograms, phylograms, rooted versus unrooted trees and that there's a bunch of possible trees for even a small number of species. So I want to turn my attention to estimating phylogenies and the most important concept here is that we're gonna be comparing features of organisms and the features we compare across different organisms are supposed to be homologous. So we went over homology in lecture one or two, but basically this is similarity in form that's caused by a common ancestry and the classic example, for instance, is the forelimb across different tetrapods where you have this pattern, the pattern of similarity is one bone followed by two bones, followed by a bunch of little bones followed by some long bones, right? Your finger bones. And so the idea is that the similarity in different tetrapods organisms with forelimbs is caused by the fact that they shared a common ancestor that had forelimbs and had that pattern of bones in the forelimb. So the idea is we're gonna study homologous characters and specifically we're gonna compare the details of these homologous characters. Now classically, these phylogenies were made using morphological traits, traits you can see with your eye or observe with a microscope. Today the homologous traits tend to be DNA sequences and specifically DNA sequences from the same gene sampled from different species. So for instance you can take the beta-globin gene and compare it in humans and chimpanzees and frogs and birds. You can actually compare that gene and that gene is homologous across those different species. The ancestor of all those species had a beta-globin gene. And so we'll give you an example. The goal is to construct a character matrix. If it's a character matrix of DNA sequences, they don't call it a character matrix, they call it an alignment. I should write that down. So if a character matrix of DNA sequences is called, you'll never hear them, people say my character matrix of DNA sequences, they call that an alignment. So let's give an example or go over an example of that. A screen down, blanket. So here's our hypothetical example we're just gonna work through quickly. And so some weird group of organisms, I was kind of hoping that the Mars exploration vehicles would have come up with some more interesting results. Basically Mars looks like it's just a bunch of dirt and rocks as far as I can tell. But the hope, my hope was that when they got there they'd have these little cameras on the Mars rovers. And you see these little creatures scurrying by, that would have been cool, but that's not what happened. But let's imagine that that did happen and that the people saw different creatures scurry by and they could see that there's different forms of these creatures. And there's the different creatures that scurried by on Mars. It has some species that had, they look like some sort of umbrella. They had these umbrellas with a head and an eye on them and the head and eye and polka dots and so forth. Those are the different species. And if you were to make a character matrix, the idea would be you'd compare homologous traits. Now we obviously can't study the anatomy of these guys too carefully. So we just have to assume that these species are all related to one another by a common ancestor and that when you see legs, that means the common ancestor of D, E and F probably had legs. So I'm just gonna start things off and say that one of our characteristics can be legs and we can score the presence or absence of legs. Either you have legs or you don't. So I would say A, B and C don't have them and D, E and F do. So often you'll have these character matrices coded by numbers, often zeros and ones. We could put words in here. No legs, legs if we'd wanted. What's another characteristic? Arms. There, some of these guys have arms, some don't. Looks like E and F have arms, right? And the other guys don't. Any other characteristics you want to put up there? Dots, okay, dots, polka dots. So it looks like A and B don't have dots, but the other four do. How about antenna, okay? So it looks like C, D, E and F have antenna. There's our antenna, and it's C, D, E and F again. Although frankly it's not clear how to score this one because to have antenna it looks like you have to have a head, but okay. What else? Polka dots. Oh, these little neck stripes look interesting. E and F have neck stripes. Once again, some people would quibble with our scoring because you have to have a neck before you can have neck stripes, and A doesn't. A head, hey that's a good one, head. Presence or absence of a head. So it looks like that guy doesn't, but everybody else, two, three, four, five. So there's an example of a character matrix. We ran out of board and frankly this is getting a little bit boring for me. There is our character matrix. Now there's a method, so I'm gonna put the board back up again so we can have some more screen or more board space. With six species we have what 105 possible trees. And so what we have to do is have a way of ordering or ranking every one of those trees according to this character matrix. And there's many methods of doing that. I'm gonna tell you a method that's not necessarily my favorite method, but it's one that's easy to describe. And this is the parsimony method. And I believe it's the one that you guys are being taught in lab as well, so this is all very consistent with what you're being heard, but what you've heard. So this is the parsimony method. And the basic idea here is we're gonna score these characters on a tree. So let's just pick a tree, I'm gonna pick a good tree. We'll put E and F, D, C, B, C, and A. So here's one of the unrooted trees that could explain the relationships of those species. And so what we wanna do is we wanna ask, how well does this tree explain those observations? And so the parsimony method says that for every tree, that every tree gets a score based on the minimum number of what changes on that tree that data requires. So for instance, this first characteristic, the one that has the presence or absence of legs, we can score that the minimum number of changes on this tree would be these guys don't have a change and these guys do. So you have a change from zero to one along this branch. Now that's not the only way we could have had the evolution of that characteristic. For instance, we could have had three independent changes all from zero to one along each of these branches for the emergence of legs. The parsimony method says no, when we have these two choices or we have many different ways we could have mapped that character onto the tree, we choose the mapping that requires the minimum number of changes. And so we would not consider this a reasonable mapping, at least the parsimony method, wouldn't consider that to be a reasonable mapping for that character. And instead, it would say we'll just map it right there, zero to one change along that branch. So I'm going to erase the zero to one change, hopefully that's understood. I'm just gonna put an indicator that this was where character one changed. Now what about character two? That trait would be a single change that occurred along that branch for character two, the arms, the dots appear here. There's a change from not having dots on this side of the tree to having dots on that side. Antenna, same place, that's character four. Character five is the stripes on the neck, E and F have the stripes on the neck, so that's character five. And character six was the presence or absence of a head. That arose right there and we'll call that character six. The parsimony method says you do this mapping exercise for every one of the characters in your character matrix, and you count them up. So we have one, two, three, four, five, six. So this tree gets a score of six. Now in principle what you would do is you could look at the other 104 trees and you find which of those trees has the minimum number of changes, requires the minimum number of changes. And according to the parsimony method, that tree is the best one. Let's go ahead and do that. Green down, blank screen. I'm gonna make a quick character matrix here for a computer program. One, two, three, four, five, six. So this is a file format that you guys are not responsible to know. It's just a file format I happen to know pretty well. It's just a text file and we're gonna put this character matrix into the text file. So I'm gonna need a little bit of your help here. So A is all zeros. One, two, three, four, five, six. B is all zeros except for last one. Z is zero, zero, one, one, zero, one. Zero, zero, one, one, zero, zero, one. I must have too many zeros up here. See, that's why it's really hard to do this in front of 700 people or whatever it is. The next one is one, zero. Thank you. And then the next two are easy. Okay, now, oops. This is gonna be our text file. I think we need to make this a unix file here. So what I'm gonna do now is we're going to, this is just the Mac terminal. I'm gonna run a program that a lot of people use for S-mini-file-loginy and I'm going to execute that or read into computer memory that file. Good. And we're going to search for the best tree. So I'm gonna ask the computer program to find the best tree. And what it did is it said, look, I went through, remember I told you the combinatorics? There's 105 possible trees. It went through, it looked at every one of the 105 trees. It found one tree that had a score of six, which should look familiar because we said that tree there had the score of six. Score of the worst tree is 12, we're not interested in that one. The number of trees it found that had a score of six, there was just one of them. And it took, well, it didn't really take zero seconds, but it didn't take very much time for it to do that, right? The computers are fast. So what does that tree look like? That's what the tree looks like, okay? It's basically the tree I drew on the board. That was a really simple matrix. I kind of knew that would be the best tree. But this computer program confirms that of all the 105 possible trees, that was the best one, and it had a score of six, yes. It means remember you're mapping each characteristic onto the tree, and you're mapping these characteristics on the tree in such a way that each one requires the minimum number of changes for all the characters. So we basically, for this particular tree, every one of these characteristics can be mapped onto it such that only one change is required. All the other trees require some of these changes to have more than one change along them. Yes, sure. We can look at a different tree, I suppose. Let's see here. Let's just take a tree that we had before was something like A, B, C, D, we'll do D, C, E, F. And which characteristic should we look at? Let's look at the evolution of legs, all right? So I'm gonna say that this guy has legs, this guy has legs, this guy has legs, this guy doesn't have legs, this guy doesn't have legs, that guy doesn't have legs, right? I just put the characteristics on the tips of the tree. So now we've got a situation where we can't, before I could put a change there, that's not gonna work. So you have either this situation where you have, let's see here, you could go from zero to one here and zero to one here for the leg characteristic, or, so that requires two changes, right? I think there's another equivalent way of mapping this one. Or you can have a change here from one to zero and a change here from one to zero. But it's, again, it requires two changes. So either of those mappings would be equivalent under the parsimony method, but it doesn't matter because they both require two changes. So this tree where it's gonna require at least a little bit, it's gonna have a longer tree length. Instead of having six, it's gonna be seven or eight. I don't know what it would be. Is that clear? This is just an example of, this is one of the other 105 trees. I showed you one of them, the best tree over there. This is one of the other 104 trees. And this is how you'd map just one of those characteristics on it. Now this is kind of a silly example. Let's not play with this anymore. Let's look at another example. So here is, here's a real alignment. I just realized I was looking where this, that's been on my hard drive for quite a while. I mean, this is a alignment of DNA sequences that was put together before most of you were probably born. But basically, this is where the paper from this Hayasaka et al. 1988, what they did is they sequenced specific genes from the mitochondrion in these primates. There's human, chimp, gorilla. There's the orangutan. There's the gibbon, some macaques, and some other more primitive primates. So this is the alignment. Note that there's 12 species in this alignment. So instead of having six like we had, we have 12 and there's roughly 654 million possible trees for 12 species. And you can see that these DNA sequences have this, they're quite long or they can be. So here, we're scrolling along. There's the end of the alignment. But basically, these are the same gene sequence from the different species. And you can see that they can be quite similar in some places. But then occasionally, like here's a case where these four great apes have a C at that position, but the other guys have a T. So let's have pop read that data set in. So execute primates.next, yeah. So I'm gonna do a search for the best tree. Now, this is the score of the best tree. The best tree required 1,153 changes on it. So instead of six changes like the best tree before, this one requires many more changes. And there were two trees that were equally good under the parsimony criterion. That is to say, of those 654 million trees, two of them had a length of 1,153. They're equally good under this method. So let's go see what those trees look like. So here is the first tree rooted at the lemur. And we'll just concentrate on this part up here because this is the only place that really, this is the, these are the great apes. So it got this part of the tree correct. And this tree puts human and chimp as each other's closest relatives with grill outside it. The other tree it found was identical in all respects, except it put the chimp with the gorilla. So this one says that the gorilla and the chimp reach other's closest relatives. This actually, this tree is reflective of the uncertainty that was in the field for a long time about whether the chimp went with the human or the chimp went with the gorilla. And this was one of the early datasets that tried to resolve this problem. Since then we've got entire genome sequences, right? Not just 898 sequences, but we have the entire genome of these critters. And we now, we now strongly believe that the human and the chimp are each other's closest relatives. But this was a good historical example as well. But this is more reflective of how people really estimate these trees today. They take a DNA sequence alignment and they feed it to one of these computer programs. This is a program that was written by a fellow David Swofford who's at Duke. And then these computer programs use computational methods to very quickly try to find the very best tree. Are there any questions about this part of the lecture which is, I erased this, didn't I? I wanted to talk, this is where I was talking about estimating phylogenies. Now there's some, oh go ahead. Which, yes. Well, so, no. I mean, what can happen for some of these analyses is they can take a very long time. Sometimes these labs have these programs running for months at a time. Now that said, if you have 200 sequences, we're just gonna agree that no computer can go through and enumerate all those possible trees. Because there's 10 to the 60 is the number of atoms and you have 10 to the 300 possible trees. And so you could wait trillions of years and never actually go through them all. That said, the methods that people use to actually find trees don't require that you enumerate all the trees. There's fast ways, they're called hill climbing methods, that basically don't look at all the trees but can come to very good solutions. Any other, that's a very good question. The question, by the way, was if you have a big data set doesn't your computer just blow up and melt on your desk and the answer is no. Okay. There's a little bit of terminology I wanna talk about which goes with the parsimony method. And that involves these terms called snapomorphy, for instance, I'll just give you some examples. So when you map a characteristic on a tree, so here we're gonna have our tree, these are gonna be, this is gonna be a dog, there's a cat, there's a human, there's a pig, say. And then over here we have a frog, now I'm not gonna put a frog there, we'll put a lizard and a snake, okay? So there's just a tree, it's probably reasonable, I don't know the details up here, but it's probably a reasonable tree. Sometimes when you map, go through this exercise of mapping characteristics on the tree, for instance, mammals are all characterized by having hair, right? And so you can say that hair evolved right there on the tree, okay? So when you look at the different characteristics in mammals, you can say that hair is what's called a snapomorphy for mammals, it's what's called a shared derived character, this is a characteristic that's shared across all mammals and derived uniquely for mammals. Now you can also say look, look at primates, monkeys and apes, right? So things that are over here, they have hair, right? So isn't hair a snapomorphy for primates? The answer is no, that's a plesium morphy, what's called a plesium morphy, plesium morphy for primates. So hair is a snapomorphy at one level, at the level of all mammals, it's shared among all mammals and derived uniquely at the base of mammals, but it's a plesium morphy for groups such as rodents or primates. Sure rodents have hair, sure primates have hair, but it wasn't a characteristic that was derived uniquely within rodents or in primates. Just a shared primitive character, shared primitive. So this is a concept that's sometimes difficult to see, but characters once again, I'll just repeat it, some characteristics can be snapomorphies at one level, but plesium morphies on another. The hair is a good example of plesium morphic at the level of any group within mammals, rodents, primates, bats, whatever, but it's a snapomorphy for all mammals. It's a characteristic that all mammals share. Other characteristics, for instance, that all mammals share are things like giving milk to a young, right? All mammals do that. Or having a very specific jaw joint in the jaw, so specific bones contact one another in the jaw, the denturine squamosal bones. So there's a lot of snapomorphies that all characteristics that all mammals share and all those characteristics are called snapomorphies. Is there a question about that terminology? You probably already got that in lab, I believe. No? Well, now you got it. Okay, how are we doing on time? We still have some time, good. Is there anything else I wanted to talk about for these methods? Okay, so what can you do with phylogenies once you have them? So there's a lot of interest among biologists and evolutionary biologists specifically to make phylogenies. And there's, for instance, a project called the Tree of Life Project, which is being funded by the National Science Foundation, which has as its goal the trying to piece together the entire Tree of Life from bacteria through plants and all mammals and vertebrates. Okay, so there's a very ambitious project out there where people are collaborating and trying to piece together the Tree of Life. And so to some extent, phylogenies are interesting in and of themselves. And I'll give you some examples at the end of this lecture or the beginning of the next of cool phylogenies, neat results that have come out. However, there's other reasons that people are interested in phylogenies. I'm gonna give you what I consider one of the more boring reasons first, and then I'll give you some more interesting examples later. And what I consider, some people in the museums here would kill me for saying this, but one of the more boring reasons is for classification. We need to be able to classify organisms to make sense of the diversity of life. And we already went through the Linnaean hierarchy, right? The kingdom phylum, class order, and so forth that is the traditional way in which species are classified. Every species gets its own kingdom, its own phylum, class, order, family, genus, and species. That's how species are classified. There is a group of very vociferous biologists out there who believe that the classification should reflect the phylogeny. So if you have a phylogeny like this, where all these things here are mammals, to these people, and to me frankly as well, a good classification would reflect this, that maybe what you should do then is call all these things a name. Give all the things that we call mammals a name. We'll call mammals, all right? Mamelia would be what they do. Now these people get upset when you have groupings that don't reflect the phylogeny. So let's give a good example. There's a very, I think a good example of a group that these people would not be happy with, a group that these people would not be happy with. And so I'm gonna sketch out what I remember of the vertebrate tree of life. So I think you have, let's see here, you have frogs here, amphibians. You have mammals. You have turtles. You have lizards and snakes. You have crocodiles. And you have birds. Which of those things are reptiles? Well, that's not, that's an amphibian. Those aren't reptiles. We would call a turtle a reptile, yes. Lizards and snakes, certainly. Crocodiles, birds, no. So if we were to make a grouping of reptilia, classically, reptilia means those organisms. Note that this classification includes the common ancestor of all reptiles. It includes this ancestor, and if you were to look at that ancestor in the fossil record, people would call it a reptile, okay? So it includes the, this grouping includes the ancestor of all reptiles, but not all of the descendants, because we don't call birds reptiles. So this is what's called a paraphyletic group. This includes the ancestor and some, but not all, but not, don't be too touchy. But not all descendants, okay? That's what's called a paraphyletic group. This other group where you have an ancestor in all the descendants, that's called monophiletic. This is a grouping that includes the ancestor in all of the descendants. And these people, biologists today, most of them believe that classifications should be only for monophiletic groups. They should only include monophiletic groups. So there are vertebrate biologists who have redefined reptilia to include birds, and that fixes the problem, okay? Although, frankly, you'd have to kind of know who you are talking to. Does that person across the way believe that birds are reptiles when you say the word reptiles, so you have the same meaning of the word? Okay, so I'm gonna continue with the phylogeny lecture next time.