Phylogenetics and Genome Evolution: Study with Video Lessons, Practice Problems & Examples

Phylogenetic trees are constructed using homologous traits, indicating shared ancestry. Genes can be homologous as orthologs, arising from speciation events, or as paralogs, resulting from gene duplication within the same genome. The molecular clock estimates divergence dates based on mutation rates, which are generally constant over time. However, variations in mutation rates across lineages and the influence of natural selection can affect accuracy. Understanding these concepts is crucial for studying evolutionary relationships and genetic variation.

Homologs: Orthologs & Paralogs

We've been building phylogenetic trees. We've been saying that we want to use homologous traits or homologous characters to build those trees because we understand homologous traits tell us that these organisms are more closely related because they share a common ancestor and that trait was present in that common ancestor. Well, when we actually look at DNA and look at genes, there's actually more than one way that genes can be homologous. Right? So when genes are homologous, we call them homologues.

So 2 homologous genes are homologues, and homologs could be homologous either because they are orthologs or paralogs. Now these are obviously similar and somewhat confusing definitions that are easy to get mixed up, so we're going to go through these now. Now let's just redefine a homologue here. A homologue, these are genes that are scented from the same ancestral gene. Alright.

Now the first way that you could be a homologue or that a gene could be a homologue, we are going to call an ortholog. And this is what we've sort of been talking about so far when we've been building phylogenetic trees. Orthologs are homologous genes in related species, and we're going to say that they arise through speciation. So an example of this, well, we're going to look at human and chimpanzee. I'm just going to write chimp though.

Human and chimp beta hemoglobin. Now just to be clear though, virtually every gene in humans has a homologue in chimps because we are very closely related species. Alright. Now let's see how this works though just looking at this image. We have our chimp and our human here, and we have a very simple phylogenetic tree here.

Right? We can trace the lineages back. They meet at a common ancestor. And what we see is that in this common ancestor, there was a version of this gene in the DNA called beta hemoglobin. Now since the speciation event, chimpanzees have a version of beta hemoglobin, and humans have a version of beta hemoglobin.

Now those aren't identical. They are very very very similar, but they're not identical because they have been evolving separately for millions of years. But because they come from that common ancestor and they are different genes just because they are in different species, we call them orthologs. Alright. Now, typically, when you're building a phylogenetic tree, you want to be using orthologs to do that.

Now, orthologs, to remember this, we're going to have a little memory tool here and it says orthologs are in other species. So just remember that, ortholog, other, and remember, it comes from that speciation event. Okay. Now, the other way that you could have a homologue, and this is the one that might be a little new to you, is a paralog. And a paralog, this is a homologous gene that is in the same genome.

Alright. So how does that work? Well, you could have genes that arise through gene duplication. Remember, we said that one of the ways that you get new genes is that a section of the chromosome gets duplicated, and now where you had 1 gene, you have 2, and that second copy can now evolve to do some new or related task. So when that happens and especially when that happens repeatedly through many duplications of these genes, you end up with what we call gene families or a gene family here.

A gene family is going to be a group of paralogs that are all in the same genome. Alright. So for example, we talked about beta hemoglobin above. Beta hemoglobin is actually one of 5 globins in the beta hemoglobin gene family. Alright.

So we can look at, here a section of DNA. This is in humans is actually on chromosome 11. There are 5 globin genes right in a row that all arose through gene duplication events. Now instead of going from the top, down, I'm going to start at the root of this tree here. Right?

So the root of this tree, this tells us that, you know, deep in the past on this chromosome, there was a single globin gene. And then a piece of that chromosome got duplicated and then there were two versions of that globin gene. Alright. We can follow this, part of the tree on the left here. Now we see that this version of that gene got duplicated again.

Right? And one of them is what we now call hemoglobin epsilon. The other version of that gene that got duplicated got duplicated again. Right? And that evolved into the other 2 gamma hemoglobins that we have.

Alright. Now hemoglobin epsilon and the gamma hemoglobins, those are all expressed at some time during fetal development. Now on the other side of this tree, right, there was another duplication. This version of this gene duplicated and it arose and is now what we call beta hemoglobin, and the other version is what we call delta hemoglobin. These two versions of the hemoglobin gene are expressed in adults.

Alright. But you can see here, remember, all of these genes are in the same genome. They are all in you. They were all expressed at some point in your lifetime, but we can draw a phylogenetic tree because we can trace back how they arose through these gene duplication events. Alright.

Now to keep paralogs separate from orthologs, in your head we have another memory tool here. We're going to say that paralogs are peas in a pod. Right? This group of genes, this gene family, it's all in a single genome just like peas in a pod. Alright.

With that, we've got practice to come. Check it out.

Which of the following statements about homologs are true?

I) Phylogenetic trees of different taxa can be made using orthologs but not paralogs.

II) A gene family is a group of paralogs within a genome.

III) Orthologs are homologs that are found in different species.

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

The different genes that code for the photoreceptive proteins that function in the vertebrate eye are all believed to be paralogs. Which statement below is most consistent with this idea? 

• A. The ability to detect light evolved repeatedly in different vertebrate lineages.
• B. Different photoreceptor genes are found in different species of vertebrates.
• C. Gene duplication events in photoreceptor genes have occurred whenever there has been a speciation event.
• D. The genes that code different photoreceptors all arose through gene duplication events.

The Molecular Clock

On some phylogenetic trees, you'll see that dates indicate when things last shared a common ancestor, when those splits occurred on the phylogenetic tree. So how do they get those dates? Well, you can do it using the fossil record, but that's not how it's done on most phylogenetic trees. Usually, you're using this concept of the molecular clock. Now you're probably going to be responsible for this concept.

It's unlikely that you need to actually calculate dates using a molecular clock, but you want to understand just generally how this works. So sort of fundamentally, what we're going to say here is that the number of mutations between taxa right? Number of mutations between taxa, you could say lineages or species, whatever. The number of mutations can estimate the date of divergence, or in other words, when those taxa last shared a common ancestor.

So why does this work? It works because mutations tend to enter populations at a relatively constant rate. Right? Mutations, it well, mutation is a random process, but over a long period of time, that means that randomness kind of evens out and it becomes quite predictable how often mutations are going to happen. Now we're going to say here that this is especially true for neutral mutations, and we're going to come back to that in a little bit as to why that's important.

But to explain this, I have this graph here, and you can see on the y-axis we have the number of mutations, and on the x-axis we have time. So each dot here is supposed to represent the comparison of 2 species, and you can see that if it's been not that long a time since they last diverged, well, there aren't that many mutations. And as you go up over time, right, the longer it's been since the divergence, the more and more mutations there have been. And then you can see there's sort of some scatter around the line. It's not a perfect relationship, but we can calculate this mutation rate.

And if we know that mutation rate, well, then we can take 2 species, and we can just say, well, how many mutations have there been? We can figure that out. We can sort of draw across to the line, and that'll tell us, that's about how long it's been since they shared a common ancestor. Now this all hinges on, though, us knowing that mutation rate. So how do we calculate the mutation rate?

We're going to say we can compare the DNA sequences of related species with known dates from the fossil record. Alright. So we're going to calibrate this using fossils. So we're going to look at lions and tigers here, and we're just going to imagine that we have a really good fossil record for when they last shared a common ancestor. So if we know when that last common ancestor lived pretty darn precisely from the fossil record, that gives us the time.

Well, then we just have to calculate the mutations, count the number of mutations for some section of DNA between these organisms, do some simple division, and that gives us our mutation rate. Now we can apply that mutation rate to other species that we don't have the good fossil record for. Alright. Now this works, and it works pretty darn well, but we just need to be a little bit careful here. We just want to say that molecular clocks are only estimates, and there are a few reasons here.

Mutation rates, well, they can differ in different lineages. Right? Mutations are random, so there's just going to be some random difference between different lineages. Now that you should be able to account for by just, you know, putting error bars on whatever your date is, you know, understanding that you're calculating a range, an estimate, not a precise date. But in very different lineages, the mutation rate can fundamentally differ.

Right? You don't want to use a mutation rate from bacteria to calculate the date of the split in big cats. Right? These are just very different organisms that almost certainly just have fundamentally different mutation rates. Now we also want to say that you can't calibrate your clock beyond the time of a good fossil record.

Right? You need fossils to do this. And a fossil record is only good for the last 500 million years or so, ago, back to about the Cambrian. So this just really actually doesn't work very well for deep time at all. And if you have lineages where you don't have a good fossil record, well, it's much harder to calculate that mutation rate.

And then finally, we're going to say here that natural selection can influence the clock. Right? So I'm going to go back up over here again to this idea of it being especially true for neutral mutations. That's because a neutral mutation, natural selection doesn't affect. So that new mutation, whether it sticks around in the population, that's again just going to be random chance.

Well, natural selection is going to influence that random chance, whether or not those new mutations stick around. It may make it that new mutations are more likely to stick around or much less likely to stick around, and that's going to drastically affect the rate of mutations entering your population, and that affects how you calculate the clock. So sometimes people come up with estimates that are actually quite wrong because the section of DNA they were looking at, they, for whatever reason, didn't understand that it was under strong natural selection. Alright. So we do just want to say, though, this does usually work pretty darn well, as long as you're willing to accept sort of a range of dates, and that's because mutations do enter populations at relatively constant rates over time, and we can use that to calculate those dates of divergence.

Alright. We'll practice this a little bit more. We'll see you there.

What is one challenge to trying to date a speciation event using a molecular clock?

• A. If natural selection acts on a sequence, it can affect the rate that new mutations become fixed in a population.
• B. Because mutations are random, in some lineages, mutations do not occur.
• C. New mutations are more likely to occur in non-neutral genes.
• D. Molecular clocks can only be calibrated if the gene used has a known effect on fossils.