Building Phylogenetic Trees: Study with Video Lessons, Practice Problems & Examples

Phylogenetic trees illustrate evolutionary relationships based on shared derived characters, which indicate common ancestry. To construct these trees, researchers utilize a character matrix that distinguishes between ingroups and outgroups, helping identify ancestral versus derived traits. The principle of parsimony suggests that the simplest explanation, or the tree with the fewest evolutionary changes, is most likely correct. Advanced methods like maximum likelihood further refine this process by considering the probability of specific character state changes, enhancing the accuracy of evolutionary interpretations.

Shared Derived Characters

We've been practicing reading phylogenetic trees, but now we want to take a look at how you actually build them. Right? How do we know that two organisms are linked together on a phylogenetic tree? How do we infer this complex branching pattern connecting all these branches at different nodes? Well, starting off very simply, we're just gonna say that phylogenetic trees are built by looking at traits or characters, right?

So traits or characters, we can use these two words synonymously. A trait, you know, it might be something morphological like, you know, a lump on a bone or it might be something like DNA evidence. But what we're gonna say here is that to build a tree, what we're looking for are shared characters, and that's because shared characters suggest that organisms are more closely related. If two organisms share a character, we're gonna assume that character was in its common ancestor. Alright.

So if we look at here, we though we can take these shared characters and break them into two broad groups, our shared derived characters and our shared ancestral characters. Now we're gonna define those and see which one of those we want to really focus in on when building trees, but first, I want to introduce you to our file generic tree here. We'll just be using this for an example. We're connecting here the fish, the frogs or amphibians, mice, platypus here, a bird or a pigeon there, and a crocodile. Now I just wanna be clear.

This is a tree that is already made. Right? So we are looking at this tree. What we're talking about here is how would you build this tree? How would you know to link certain things on this tree?

Alright. Well, we're gonna say you would use share derived characters. A shared derived character is a trait that is present in more than one taxon. Right? That's that shared part, and we're gonna say that it evolved within the tree.

We're gonna say that evolved since the root of this tree. That's that idea of this being a derived character. Now we're gonna say that the shared derived characters, that's what you wanna use to build trees. So for example, in mammals, how did we know to link the mouse and the platypus together? Well, they share some characters that are derived.

For example, mammary glands and hair. Right? Both of these organisms have mammary glands and hair, so they are shared. Now, are they derived? Well, did the root of the tree, the organism that was alive at the root of this tree way back there, did that organism have mammary glands and hair?

I don't think so. Right? So therefore, I can infer that the mammary glands and hair evolved in the common ancestor of the mouse and the platypus, so I would feel comfortable linking them together. In other words, I think that those traits, those characters, evolved on this branch here in my tree. Now we have to compare that or contrast that rather with shared ancestral characters.

A shared ancestral character, that is a trait that was present at the root of the tree. So for example, all of these organisms have bones. Right? They all have bones. That's a shared character.

Most of them, you know, everything except the mouse, they all lay eggs. It's another shared character. But when I look at this tree, I go back to the root. I think that that organism at this root, I think it had bones and I think it laid eggs. So that's not informative to me.

Right? It doesn't tell me that these things are more that certain ones of these organisms are more closely related to each other. It just tells me that all of them have a common ancestor, but I knew that. Right? I was gonna link them all on the tree anyway.

I knew that somewhere they have a common ancestor. So these shared ancestral characters, yeah, they're shared, but they're not helpful to me for interpreting who is more closely related to each other because they are present at the root of the tree. Now just one quick note, you may look at that egg-laying and be, hey, what about the mouse? The mouse doesn't lay eggs, So is not laying eggs, a derived trait in that mouse? Yes, it is.

But it's not gonna be helpful to us because it is not a shared derived character. In this tree, the mouse is the only thing that does not lay eggs.

Building Phylogenetic Trees Example 1

Here we want to use our knowledge of the organisms on the tree below to decide if the following characters are shared derived characters, which we will indicate with an sd, shared ancestral characters, which we will indicate with an sa, or neither, which we will indicate with an n. Alright. And the tree we have here showing relationships of many of the lineages of the vertebrates. Right? So we have sharks, we have fish or the ray fin fish there.

We have frogs or amphibians. We have mice or mammals there. We have lizards. We have birds or pigeons there, and we have crocodiles. Alright.

So let's go through these and think what we think. So what about having 4 limbs? Do you think that is a shared derived character, a shared ancestral character, or neither? Alright. Well, all of these taxa have 4 limbs, so it is shared.

Now the question is, is it a shared derived character or shared ancestral? So I go back to the root of my tree here. Do I think that this organism had 4 limbs? Well, also descended from the root of the tree are sharks and these ray fin fishes there. They don't have 4 limbs.

So I don't think that that is an ancestral trait. That means that I think having 4 limbs is a shared derived trait. I think it probably evolved in this branch right here, leading to the ancestor of what are all really the land vertebrates there. Alright, moving down, we have the presence of gills. Alright, so what do you think presence of gills?

Shared ancestral, shared derived, or neither? Alright. Well, more than one taxon. Two taxa on this tree have gills, so it's shared. Is it at the root of the tree then, is the question?

Yes, I think that the root of the tree was probably some kind of fish, so it had gills. Okay. So the presence of gills, it is shared, and I think it's ancestral because it was present at the root of the tree. Alright.

And what about having feathers? Well, birds have feathers, right? Do I think that those feathers were present at the root of the tree? I don't. So that's a derived trait, but it's not a shared derived trait in this tree.

The birds are the only taxa that have feathers, so I can say that this is neither. Again, it is derived, but it's not shared. Alright, now what about having lungs? What do you think? Alright.

Well, the same origins with 4 limbs also have lungs, so it is shared. If I go back to the root of this tree, do I think that this organism had lungs? I don't. So I think it is a shared derived trait. Now, a fun little fact, lungs actually evolved on this branch here.

They evolved before the split between the ray fin fish and the land vertebrates. This ancestor had both lungs and gills and actually, this type of fish, sort of the fish that you're most familiar with, they've actually lost their lungs. Their lungs evolved into a swim bladder. Fun fact. Alright, moving on, we have laying eggs.

So what do you think about laying eggs? Well, laying eggs is definitely shared; most things on this tree lay eggs, but is it ancestral or derived? Well, if I look at the root of the tree, I think the ancestor laid eggs, so this is shared ancestral. Alright. What about terrestrial reproduction?

In other words, reproducing on land. Alright. Well, what do you think? Well, these 4 organisms or 4 taxa there, they all reproduce on land. The other ones lay their eggs in the water, so it's definitely shared.

It is not present at the root of the tree. I think it evolved on this branch right here, so that is a shared derived trait. Next, we have producing milk. What do you think? Well, producing milk is definitely derived, not present at the root of this tree, but in this tree, it's only present in the mouse, right?

If I had other mammals there, it would be a shared trait. But I only have one taxon here that produces milk. So it is a derived trait, but it is not a shared derived trait. So I'm going to put an n for neither. And then finally, we have teeth.

What do you think about teeth? Well, I think that teeth are definitely shared. Right? Everything on this tree except the bird has teeth, but I think it's ancestral. I think that the root of this tree, the organism had teeth.

Alright. So, with that, more practice follows. I'll see you there.

Which of the following statements are true regarding shared derived characters?

I) Shared derived characters are examples of analogies. 

II) Shared derived characters are traits shared by multiple taxa within a tree. 

III) Shared derived characters are used to establish evolutionary relationships.

• A. I & II.
• B. I & III.
• C. II & III.
• D. I, II, & III.

The following table shows an inferred ancestral sequence for five DNA nucleotides. Then it shows the sequence for the same five nucleotides in taxa X, Y, & Z. If you wanted to build a phylogenetic tree for taxa X, Y, & Z, which nucleotide would be the most informative?

• A. Nucleotide 1.
• B. Nucleotide 2.
• C. Nucleotide 3.
• D. Nucleotide 4.
• E. Nucleotide 5.

Ingroups, Outgroups, & the Character Matrix

We've seen how shared derived characters are essential for establishing evolutionary relationships. But now we want to see how we actually build this tree. So, to do that, we're going to need to introduce ingroups, outgroups, and the character matrix. And what we're going to do here is we're going to go over this vocabulary first because it's very likely that you're responsible for this vocabulary on some level. So we'll go through that first, then we'll show you how we can actually use a character matrix to build a phylogenetic tree.

Now it is less likely that you actually have to do that as a skill, but I do want to show you how it works because I think following along with that will really help you understand how we actually build trees. How do we establish these evolutionary relationships? Alright. So let's get going with this vocabulary. We're going to start with the character matrix.

Character matrix so it's a matrix. So basically it's a table, and it's going to be used to identify these shared derived characters. Right? And we have a character matrix here that we're going to fill in. You can see here that going down one side, it has organisms.

And going across the top there, it has different characters or different traits. Now if you were doing morphological traits, you know, the type of traits that we have here, typically you would be using dozens of them if you could. Because remember, more characters make a better tree. And if you're using DNA as your traits, you might use 1,000 of them or as many as you can. Alright.

Now you'll notice though that I've put my taxa here going down the side in 2 groups. I've color coded them. So why are they like that? Well, I've broken my taxa into 2 groups. The first group we call the ingroup.

The ingroup are these taxa that we wish to link together or connect in our phylogenetic tree. Right? So in this case, this is the frog, the mouse, the platypus, the bird, and the crocodile. So these are the organisms that we're really trying to build this tree for. We're trying to establish how are these organisms related to each other.

Right? That's our ingroup, but then we also have an outgroup. An outgroup, we are going to choose a related taxon, a taxon that we know is more distantly related than anything in the ingroup. Alright. So I have my ingroup, and for my outgroup, I've picked a fish.

Right? So what I want to do is I want to pick something that I know is pretty closely related to the things that I'm trying to build the tree for, but I know isn't in that group. Right? So my group that I'm trying to build the tree for is land vertebrates, and so as an outgroup, I've picked the fish. Well, fish, those are not land vertebrates, but they are vertebrates.

Right? So I think they're probably pretty closely related, but they're definitely outside that group. Now the reason we use an outgroup is that this is going to help us identify those shared, derived characters. Specifically, it's going to help us identify the difference between a derived character and an ancestral character, and that is going to help us correctly place the root on our tree. Alright.

So let's see how this is going to work. So you see our characters there or see, rather, our taxa are there. You see the characters going across. We have whether or not they lay eggs, whether or not they have 4 limbs, if they have the amnion, that that layer that surrounds a developing embryo in some organisms, whether or not they have mammary glands, and whether or not they have a gizzard. This structure in some organisms that grinds up food.

Alright. Now to fill this out, all we're going to do here is we are going to write a 1 if the taxon has that character, and we are going to write a 0 if it does not. Alright. So let's do Lay eggs. Does the fish lay eggs?

Yes. So it gets a 1. What about the frog? Yes. It gets a 1.

The mouse? No. It gets a 0. Platypus, bird, crocodile, they all get ones because they all lay eggs. Alright.

Let's keep going. Four limbs. Fish does not have that, but everything else does. Alright. What about the amnion?

The fish and the frog do not have an amnion, so they get zeros, but everything else does. They get ones. Mammary glands, well, not the fish and the frog, but yes to the mouse and the platypus, and no to the bird and the crocodile. And finally, the gizzard. Well, the first four organisms here do not have a gizzard, but the bird and the crocodile do.

Great. Alright. So we filled in our matrix. So now let's see how do we use this thing. Alright.

So to do this, I'm going to sort of just clear everything out here. And, well, we're going to have some steps. Step number 1, we are going to identify those shared derived characters. And importantly, we are going to ignore all other characters. If it is not a shared derived character, we're going to ignore it from here on out.

Alright? So let's identify those. So going first, let's look at laying eggs. Well, the fish lays eggs. So the fish is our outgroup.

So if this is shared with the fish, we can assume that it was present in that common ancestor for everything in the tree. That means that it is an ancestral character, so we can ignore laying eggs. Alright. Now, you may say though, what about the mouse? The mouse doesn't lay eggs?

Isn't that therefore a derived character? Yes, it is. But the mouse is the only thing here that doesn't lay eggs. It's not a shared derived character. Remember, our goal here is to put things in groups.

It's to link things together. It doesn't tell me anything that things are different from each other. Right? These are different organisms. I know that going in.

What I want to know is what do they share with other organisms so that I can assume that they have a common ancestor with them. So, therefore, it's not a shared derived character. I can ignore laying eggs. Now everything else here, everything else is shared by more than 1 character. I'm sorry, shared by more than one taxa, but it is not present in the fish.

So all my other characters here, they are shared derived characters. Alright. We can now move to step number 2. Step number 2 says identify pairs of taxa with the most shared derived characters. These are sister taxa.

Alright. So now I got to count how many of these shared derived characters each organism shares with each other. So to do that, I'm going to make another little table here. Going down the side, I have my ingroup, and going across the top, I have my organisms that I'm going to compare my ingroup with. Alright.

So we can start with the fish. Alright. How many shared derived characters does the fish share with the frog? Well, the fish is the outgroup. So it's kind of by definition.

It shares no shared derived characters with any of these organisms. Organisms. Anything it shares with these organisms, we're going to assume is an ancestral character. That means that going down for the frog, the mouse, the platypus, the bird, and the crocodile, it can just put zeros there. It doesn't share any.

Alright. Well, then what about the frog? The frog, you can see here, it has one share derived character. It has 4 limbs. Who does it share that with?

Well, it shares that one character with the mouse, so they share 1. It shares that one character with the platypus, so they share 1. It also shares it with the bird and with the crocodile. So the frog shares 1 share derived character with each of those organisms. Alright.

We can keep going here. The mouse, the mouse has 3 share derived characters. Well, how many of those does it share with the platypus? It shares all 3, so they share 3. With the bird, it shares 2.

With the crocodile, it shares 2. Alright. We keep going. What about the platypus? It has 3 share derived characters, but, well, how many of those does it share with the bird?

Only 2. How many does it share with the crocodile? It shares 2 with the crocodile. Alright. Our last comparison here, we need to compare the bird and the crocodile.

The bird has 3 shared derived characters. How many of those does it share with the crocodile? It shares all 3. Alright. So now we need to identify the taxa with the most shared derived characters.

So what's the biggest number in this table down here? The biggest number is a 3. Now there are two places where there are threes, but let's start with this one. Alright? So first here, we see that the bird and the crocodile have 3 shared derived characters.

Between them, that's the biggest number in my table. So that tells me that I think they are Sister Taxa. They have so many shared derived characters because they share a recent common ancestor. So I'm going to start drawing my tree by joining those 2 together as a group.

Alright. Well, the platypus and the mouse, they also share 3. So again, I think they share so many because they share a recent common ancestor. So I can link these 2 together on my tree. Now these are a group.

Alright. So now I've found my sister taxa. Now, I can go to number 3. I'm just going to continue connecting the taxa based on the number of shared derived characters. But I just want to call out one thing.

I've already linked together my mouse and my platypus. I can now treat them as a group. Right? I've already linked my bird and my crocodile together. I can treat them as a group.

Alright. So with that in mind, let's look what's the next biggest number on this, on this table here. Well, it's all these twos. Alright. Now what do I get, or what comparison gives me those twos?

It's when I compare anything from those two groups together. Right? When I compare either the mouse or the platypus to either the bird or the crocodile, I get a 2. That means that's my next highest number. I can link those taxa together.

Those now make a bigger group of those 4 taxa together. Alright. So now those are linked together. Now I can keep going. What's my next highest number on this table?

It's this one. Well, this one is whenever I compare the frog to anything in this big group here. So now I can link my frog onto this tree. And the final thing I have to do, I need to link on my outgroup. Well, the outgroup, I don't care what the numbers are because by definition, I'm going to link my outgroup to the bottom of my tree, and that's going to give me my root.

Alright. So we've built a tree. That's how we do it. Alright. Now going forward, remember, ingroup, outgroup, and a character matrix, those vocabulary terms you're very likely going to be responsible for.

Actually building a tree, it's less likely you need to know how to do that. But again, I hope you followed along and hope to see how this works because, well, I think it's really cool and kind of fun. All right. We're going to have more practice coming. I'll see you there.

Building Phylogenetic Trees Example 2

Our example asks, based on the data below, which two species do you think are sister taxa? All right. So we have this character matrix here, and in our character matrix, we're looking at 10 different nucleotides here as our characters, and we have an outgroup, and then we have 3 taxa here, taxa Q, taxa R, and taxa S in our ingroup. Alright. So just remember, how do we figure out what our sister taxa?

Alright. So we have to figure out between taxa Q, R, and S, which one of those share the most derived characters. Right? Remember when I did this previously, I had a little table to do that. I'm not going to draw the table.

I'm just going to write every pairwise comparison. We have to count how many are shared between Q and R, how many are shared between Q and S, and how many are shared between R and S. Those are all the possible sharings that there could be. And now I'm just going to go through each nucleotide, decide if it is a shared derived character, and if it is, which taxa share it? And I'll just put a little tally mark next to the sort of pairwise comparisons that I have if it's shared between those two.

Alright. So let's start. Number 1, is that a shared derived character? It's not. Right?

So those characters are different from the outgroup, but none of them are shared. So I can basically just ignore nucleotide 1. What about nucleotide 2? Alright. These are shared, and they are different from the outgroup, but they're shared between all of these.

So, it's actually not very helpful. It's not going to help me make a group within my ingroup, to link anything else. Now I'll still just go ahead and put tally marks there, but I'm going to put a tally mark, Q and R, show you that, Q and S, share that, R and S share that. Again, it's not going to help me very much in the end. Alright.

But we can keep going. Nucleotide 3 here. What about nucleotide number 3? Is that a shared derived character? It is.

Right? These C's are shared, and they're different from the outgroup. So that means that I can put a tally mark next to Q and R, because they share that shared derived character, so it gets a tally mark there. Alright. What about nucleotide 4?

Yeah. Nucleotide 4, taxon R and S have that shared G character, and that's a derived character. So I'm going to put a tally mark here next to taxon R and S. They have a shared derived character between them. What about nucleotide 5?

Right. Well, those C's are a shared derived character because it's shared between Q and R, and it's different from the outgroup. So Q and R get another tally mark there. What about 6 here? Alright.

Number 6 is the same thing as nucleotide 2. They it is shared. It doesn't give me much information though, because it's shared between all of them, but I'll go ahead and put a tally mark. It doesn't hurt. All right.

We then have nucleotide 7 here. Is this a shared derived character? Well, yes, it is because the C's are different from the G, and it is shared between taxon Q and R. So Q and R now get another tally mark. They're starting to rack them up here.

What about 8? All right. Not a shared derived character. There is nothing shared about those characters. They are all different.

What about nucleotide 9 here? Well, nucleotide 9, you see that the T's are shared between Q and S, and that's different from the outgroup. So I'm going to put a tally mark next to my Q and S comparison here. And then finally, what about 10?

Oh, yeah. 10 shows me that Q and R have that shared derived character that is different from the outgroup, so it gets another tally mark. Alright. So what are Sister Taxa? Well, Sister Taxa here are the ones with the most shared derived characters, and that would be taxa Q and R.

I'm going to assume that they are a group because I'm going to assume that the majority of those shared characters shared derived characters are homologous traits, and I'm going to link them together on my phylogenetic tree. Now these other ones have shared derived characters, but remember some of these could be analogous traits. And that's why we try and get lots and lots of characters, because in the end, the most of them will be homologous, and the biggest number is going to win out. Alright.

So Q and R, that's my answer.

You are constructing a phylogenetic tree of all species in the family Bovidae, which includes all cattle, sheep, goats, antelope, bison, and buffalo. What do you think would be a good outgroup?

• A. A moose.
• B. A domestic cow.
• C. A fish.
• D. An oak tree.

You are constructing a phylogenetic tree of the oak trees, and you include one species of chestnut tree as an outgroup. Why would you include the chestnut species?

• A. To test if oak trees are a monophyletic group or if some may be more closely related to chestnut.
• B. Adding the chestnut will help establish branch lengths for the oak species.
• C. The chestnut will help establish which derived characters are homologous and which are analogous.
• D. The chestnut species will help add the correct root to the tree as it helps to establish ancestral traits.

Based on the following character matrix, which of the following organisms would you predict to be sister taxa?

• A. New and Old World monkeys.
• B. New World monkeys and apes.
• C. Lemurs and New World monkeys.
• D. Old World monkeys and apes.

Parsimony

We continue thinking about building phylogenetic trees, we now need to introduce a concept and that's because, well, we're going to say here that phylogenetic trees are built using the concept of parsimony. So parsimony, we're going to define here. Parsimony is this concept just that the simplest explanation is probably correct, right? Now this you can take outside of biology. And in general philosophy, this is often referred to just as Occam's razor.

Now, Occam was a philosopher, and what he said was you want to take your explanation and shave it down to its simplest form because, again, that simplest explanation is probably the correct one. Now when we're looking at phylogenetic trees, we can take what we call a maximum parsimony approach, and that just says that a tree that represents the fewest number of evolutionary changes is most likely. It's probably the correct tree if you had the least number of evolutionary changes, or we can think of character state changes happening on that tree. Now the way this actually works in modern biology is you'll take your data, you'll put it in a computer. It might be thousands of characters, you know, if you're looking at DNA evidence, and that computer is going to build all sorts of very similar phylogenetic trees.

And then it's gonna use the concept of parsimony to figure out which one of those very similar trees is the most likely explanation. Alright. Now, to sort of model how this is gonna work, we're gonna compare 2 very similar trees here with these characters and character states that we've looked at before. So here our trees show. Well, we have a fish, a frog, a mouse, a platypus, a pigeon, and a crocodile.

It's the same organisms we've looked at before and in the same organism in both trees here. And the difference between these trees, well, the tree on the left, we see that the mouse and the platypus are shown as more closely related to each other. Whereas, this tree on the right, we're kind of testing this hypothesis that, well, maybe the platypus is more closely related to the bird in the crocodile. All right, now figure this out. We're gonna map our character state.

So here are our character states that we're going to be looking at. So let's start, and every time we think that there should be a character state change on our tree, we're going to map it. To see how this works, I'm going to use a green dash for the ability to lay eggs. Now laying eggs is actually an ancestral trait, so I don't need to map that character state change on here because that comes before the root of this tree. But we know that mice don't lay eggs, right?

So that's also a character state change. So I can put a dash on this lineage that leads to the mouse, right? Because on that lineage, it lost the ability to lay eggs or evolve the ability to have, you know, internal fertilization, whatever you want to say. Now on this tree over here, same thing, lineage leading to mice. I'll put this dash because we have a character state change.

The mouse doesn't lay eggs. Alright. Now we can keep on going for all of our other traits here. We have 4 limbs. I'll mark that one in red.

Well, the ancestor of all 5 of these organisms on both trees, that's where we think that 4 limbs evolved. So I can put that character state change on that branch that leads those 5 organisms. We have the amnion. The amnion, I'll mark in this sort of yellow, orange, or gold color here. Remember, the amnion is this layer that surrounds the embryo in these 4 organisms.

So I find the branch that leads to those 4 organisms, and I put that character state change on my tree. I'm gonna do the gizzard. Gizzard here in this blue. Gizzard, the structure that grinds up food and the bird and the crocodile, so it would have evolved in this ancestor that leads to those organisms on this tree. And then we have the mammary glands.

The mammary glands, these milk-producing organs. Now on the left, that's present in the mouse and platypus, so I find the branch that leads to them, and I have the character state change where we gain the ability to make milk, gain the mammary glands there. But then on this tree over here, well, I have the mouse and the platypus, but they don't share an immediate common ancestor with each other. So how is this going to work? Well, now I could say that maybe the mammary glands evolved independently in each organism, right?

That's one possible explanation. That would be 2 character state changes. Now another explanation, that's what I'm gonna draw in here. It's still gonna be the same number of character state changes. It's gonna be 2.

We could have evolved mammary glands in the ancestor of all of these organisms, And then maybe mammary glands were lost in the lineage leading to the bird and the crocodile. Again, not saying this is what happened. We're just testing this hypothesis. Okay, so that's a possibility. So which tree is the correct tree then?

Or which do I think is the correct tree? Well, I'm gonna count my number of evolutionary changes. Well, this one on the left, I see 1, 2, 3, 4, 5 character state changes. So I'm going to put a 5 here, 5 evolutionary changes. My tree on the right, 1, 2, 3, 4, 5, 6, 6 evolutionary changes.

Alright. Therefore, by the concept of parsimony, I can reject this tree. I think the tree on the left there is more likely to be the correct tree because it is more parsimonious. There are fewer evolutionary changes. Alright.

Now in modern biology, we still use the concept of parsimony, but we understand that things can be a little more complex than just counting the number of changes. So we use more advanced statistical methods, things like maximum likelihood. Now maximum likelihood sort of builds into our model and just sort of recognizes that some changes we're going to say are more likely than others, right? So it's not just as simple as counting the number of changes. We want to make sure that the more likely changes are happening more often.

Now, for example, we know that certain mutations. We know that certain mutations are more likely than others. So for example, a mutation from a cytosine to a thymine, we know that is more likely than a mutation from a cytosine to a guanine, from a C to a G. And that's just due to the chemistry and the structure of DNA. So we can build that into our model and say, what we actually think is more parsimonious might be a few more evolutionary changes if those evolutionary changes are more likely to happen. So again, remember the idea of parsimony, go with the simplest explanation, and you can take this again outside of biology.

If you're trying to win an argument with a friend, you've got a new 50¢ word to use. Alright. More after this. I'll see you there.

Building Phylogenetic Trees Example 3

Example tells us that in the tree below, the tinamous are the only birds that can fly. All other birds on this tree are flightless. Now to analyze this tree, you will consider the character of flight. There are 2 character states for flight, flighted and flightless. All right, so we see this tree of these birds.

Now we said before, these are mostly large birds that cannot fly. However, now we know that the tinamou here, this is a bird that can fly. Tinamous are a group of birds that actually live in South America. Alright. So now knowing that and looking at this tree, let's answer our questions here.

A says, circle the ancestral state that would make the most parsimonious tree possible. Was that ancestral state flighted or flightless? Alright. So looking at this tree using strict application of parsimony, what do you think was happening at the root of this tree? Well, almost all of these birds, except for the Tinamou, can't fly.

Right? So the most parsimonious explanation is probably that the ancestor was flightless. Right? And if you say that that is what we think happened, how many evolutionary changes would we have on this tree with that character state of flight? Well, we would just have one.

I'll map it on here. Right? We would have the Tinamou, The Tinamou’s ancestor there that would gain the ability to fly. Right? That's the most strictly parsimonious explanation.

And, well, we'll say here that's one number of changes. 1, one change. Alright. C then says, however, assume that tinamous are the only birds that retained the ancestral state. Alright.

So now we're saying the root of this tree, we think, could fly. And that ability to fly was retained by the ancestors of the tinamou. So, therefore, what is the most parsimonious explanation for how many evolutionary transitions occurred on this tree for the character of flight? Alright. Now to figure this out well, take a look.

What do you think? So to figure this out, I'm going to sort of draw on here all the branches where I think birds could fly. Right? So it was retained from the ancestor, and then I have to trace this up all the way until I get to the. Right?

Because those retain that ancestral state. Alright. So now knowing that, how many evolutionary changes do I have? Well, if I follow the ostrich back, its ancestor, it could fly. Right?

So I gotta put a character state change there. If I follow the Ria back, well, its ancestor, it could fly. Right? So I need a character state change there because the Ria can't fly. If I follow the Kiwi back, well, its ancestor could fly.

Right? So I need a character state change on the lineage leading to the Kiwi. And then if I follow the cassowary and the emu back, well, they share a common ancestor. And as I follow that back, we get to that ancestor that could fly. So I think I need a character state change on the lineage leading to the cassowary and the emu.

Alright. So now that gets me 1, 2, 3, 4 evolutionary changes, and that's actually what we think happened. So again, just to recap everything that we've seen here, under strict parsimony, we might assume that the Tinamou gained flight and the ancestor on this tree was a flightless bird. But flight is a complex trait, and it's really unlikely that birds descended from things that could fly. Well, their ancestors of this group lost the ability to fly and then some of them got it back.

We think that is really unlikely. So what we think is actually the most parsimonious explanation is that the Tinamous retained their ability to fly, and we have 4 evolutionary changes, whereas flight was lost independently 4 times in this group. Now that may not be what you expected and actually caused some people to make this tree incorrectly for a very long time, but this is now what we think is the most parsimonious explanation. Alright, with that more practice to come. Give it a try.

What does it mean to apply the concept of parsimony to an evolutionary tree?

• A. A tree should consider both the number of evolutionary steps and the likelihood of each step.
• B. A tree should clearly display which evolutionary changes occurred within each lineage.
• C. A tree that has the fewest evolutionary changes is usually the correct interpretation.
• D. A tree should have as few branch points (nodes) as possible.