Welcome to Last Minute Lecture.
This free chapter overview is designed to help students review and understand key concepts.
These summaries supplement not replaced the original textbook and may not be redistributed or resold.
For complete coverage, always consult the official text.
Welcome back to The Deep Dive.
Today we are taking on a pretty big topic, the synthesis of
and Mendel's genetics, you know, what we call neo -Darwinism.
Exactly.
It's that moment where evolution gets its, well, its mathematical legs, you could say.
It provides the framework to actually quantify how life changes over time.
And we're looking at this on two levels, right?
That's right.
We've got microevolution, the smaller shifts, changes in gene frequencies within a population.
And then there's macroevolution, the big picture stuff like how new species actually emerge over, you know, geological time.
So our mission today for you listening is to walk through the core tools of population genetics.
These are the ideas that connect those small changes to the grand sweep of evolution.
And we're starting with something maybe a bit surprising.
Which is just how much genetic variation is actually out there hidden within almost every population.
Yeah.
When you think about a gene pool, all the genes in a group of interbreeding individuals, you might kind of assume selection has, I don't know, cleaned things up, favored the best version.
You might think that, but the potential for variation is just enormous.
I mean, you don't even need a microscope for some examples.
Like dogs.
Dogs are a perfect example.
Think about Chihuahuas and Great Danes.
They come from the same ancestral wolf stock.
That incredible range had to be lurking in the original gene pool, just waiting for artificial selection to pull it out.
Okay.
That's the visible stuff, the phenotype.
But what about the variation we can't see at the DNA level?
Ah, yes.
That's where molecular studies really blew things open.
Martin Kreitman's work on the ad gene in fruit flies, Drosophila, back in the day, was foundational.
What did he find?
He looked at this one gene across different fly populations and across about, what, 2 ,700 base pairs, he found 43 different nucleotide positions that varied.
43?
That sounds like a lot for one gene.
Wouldn't that mess things up?
Here's the really interesting part.
Of those 43 differences, only one single change actually led to a different amino acid in the protein the gene codes for.
Only one?
Seriously?
Yep.
The other 42,
they were either in parts of the gene that don't code for protein,
or they were synonymous changes, different DNA code, but it still makes the same amino acid.
It showed this huge amount of silent variation that selection doesn't even act on directly.
Wow.
And scaling that up to humans,
the numbers must be astronomical.
Yo, completely.
The 1000 Genomes Project, for instance, cataloged something like
over 88 million genetic variants in human populations.
SMPs, insertions, deletions, it's just a staggering amount of raw material.
So, okay, the big question then is why?
If natural selection is supposed to optimize things, why is all this variation just hanging around?
That's been a central debate.
Part of it is selection.
Sometimes, like with the sickle cell trait, giving heterozygotes an advantage against malaria, selection actively maintains different alleles.
But the other really key idea is the neutral theory of molecular evolution, proposed by Motu Kimura.
The neutral theory.
That sounds counterintuitive.
Doesn't selection drive everything?
Well, Kimura argued that most of the mutations that actually spread through a population might be functionally neutral.
Their fate isn't decided by advantage or disadvantage, but just by random chance genetic drift and how often they pop up mutation rate.
So it provides a baseline.
Like, assume it's neutral unless proven otherwise.
Exactly.
It's the null hypothesis.
Biologists need actual evidence for selection before they can rule out the simpler explanation of just random mutation and drift causing the patterns we see.
So if we want to measure evolutionary change, we first need a benchmark, a model of a population that's not evolving at all.
And that's the Hardy -Weinberg Law.
The famous 2p $ plus 2pq plus q2 equals a 1.
That's the one.
It describes the mathematical relationship between allele frequencies, p and q, and genotype frequencies, p squared, 2tq, q squared, in a perfectly stable, non -evolving population.
It's like the theoretical ideal, right?
But for it to hold true, you need a whole list of conditions.
Basically,
nothing interesting can be happening.
Pretty much.
There are five key assumptions.
One, everyone survives and reproduces equally selection.
Two, no new alleles appear, no mutation.
Three, nobody moves in or out, no migration.
Four, the population has to be infinitely large so random chance doesn't mess things up.
And five, mating has to be completely random.
If all those hold, the population stays in equilibrium.
Let's make that concrete.
So we have two alleles, a and a.
Frequency of a, let's call it p, is 0 .7.
Frequency of a q is 0 .3.
Okay.
So using the Hardy -Weinberg equation, the frequency of AA genotypes next generation will be 222, so $2 .7 times 0 .7 times 0 .491, or 49%.
Right.
And heterozygous, AA?
That's 2pq, so $2 times 0 .7 times 0 .422, 42%.
And homozygous is 2 .3 times 0 .3 times 0 .3 times 29%, adds up to 100%.
Exactly.
And the crucial part is, if you calculate the allele frequencies from those genotype frequencies, 49 % IA, 42 % IA, 9 % p, you get back exactly p .7 and q .3.
The frequencies don't change.
Equilibrium.
Okay.
That's the ideal math.
But how useful is this in the real world, say, for human genetics?
Hugely useful.
Sometimes we can directly count alleles.
Like with the CCR5 gene, there's a deletion, A8O32, that gives some HIV resistance.
Researchers can actually genotype people, find out if they're 11, 132, or 80 to 32, 80 to 32, and just count the alleles in their sample to find the real world frequencies.
But what about when you can't genotype, everyone, especially for rare conditions?
Ah, that's where Hardy -Weinberg really shines.
Think about cystic fibrosis.
It's a recessive disorder.
The frequency of affected individuals who are homozygous recessive genotype A is about 1 in 2 ,500 births.
Okay.
So that's 2 killed 2.
2 tooth equals $1 ,200.
Right.
Which is 0 .0004.
So we can just take the square root to find the frequency of the recessive allele q.
Square root of 0 .0004 is 0 .02.
So q is 0 .02.
Exactly.
And if q is 0 .02, then p, the frequency of the dominant normal allele, must be 1 .0 to its 0 .98.
And now we can calculate the frequency of carriers at the heterozygous.
Two p equals do it.
Let's do it.
2 to the p times q times q times 0 .98 times 0 .02.
That comes out about 0 .04 or 4%.
Wow.
So even though the disease only affects 1 in 2 ,500 people, about 1 in 25 people are carriers.
That's the power of Hardy -Weinberg.
It reveals how most rare recessive alleles are actually hiding out, protected, in heterozygous.
It's a fundamental insight.
And the law can be extended too for things like multiple alleles like ABO blood groups or X -linked traits.
Okay.
So Hardy -Weinberg is the no -change baseline.
Now let's break those assumptions.
What happens when things do change?
What are the engines of microevolution?
Well, the big one, the one Darwin focused on, is natural selection.
This happens when different genotypes just have different rates of survival or reproduction.
Their genetic contribution to the future isn't equal.
We measure that contribution with fitness, right?
Usually symbolized as W.
That's the term, yes.
And selection can push populations in different directions.
Like directional selection, where one extreme is favored.
Exactly.
The classic study is Peter and Rosemary Grant's work on Galapagos finches.
During a drought, finches with bigger, tougher beaks were the only ones who could crack the hard seeds left.
So the average beak size in the population shifted towards that larger extreme.
Makes sense.
But selection can also keep things the same, can't it?
It can.
That's stabilizing selection.
It acts against both extremes.
Human birth weight is the textbook example.
Babies that are very small or very large have higher mortality rates than those around the average, about 7 .5 pounds.
So selection keeps the population clustered around that intermediate optimal.
Okay.
And the third type.
Disruptive selection.
This is less common, but it favors both extremes over the intermediate It can potentially split a population into two distinct forms.
Lab experiments on fruit fly bristle number have shown this, creating populations with either very high or very low numbers, but few in between.
All right.
So selection is key.
What else breaks Hardy Weinberg?
Mutation.
This is where totally new alleles come from.
It's the ultimate source of all variation.
But by itself, mutation rates are usually quite low.
Think contraplasia, a form of dwarfism.
The mutation rate is tiny, like a $1 .4 times 10 fighter.
So mutation alone doesn't change allele frequencies much in large populations.
Not quickly anyway.
Got it.
What about movement?
Migration, or more technically gene flow.
When individuals move between populations and interbreed, they carry their alleles with them.
The main effect is that it makes populations more genetically similar to each other.
It reduces differences.
Any examples of that?
Yeah.
Look at the IBL blood group in humans across Eurasia.
Its frequency is highest in Central Asia, and then gradually decreases as you move west toward Spain.
That cline is thought to be a genetic echo of historical migrations, like those of the Mongols centuries ago.
Interesting.
Okay.
What's next?
Genetic drift.
This is all about random chance.
Allel frequencies fluctuating unpredictably, just because of sampling error from one generation to the next.
It's only really a powerful force in small populations.
And how does drift happen?
Two main scenarios.
The founder effect.
A new population is started by just a few individuals.
Whatever alleles they happen to carry, even rare ones, might become common in the new population just by chance.
Like the example in the book, oculocutaneous albinism in the Navajo.
Precisely.
A specific large deletion causing this form of albinism is much more common in the Navajo than in related groups like the Apache or in the ancestral Asian populations.
The most likely explanation is that the allele was present, maybe just by chance, in one or a few founders when the Navajo population became distinct hundreds of years ago.
The other mechanism is a genetic bottleneck, where a population crashes to a small size, and the survivor's alleles are just a random subset of the original variation.
Okay, one assumption left.
Random mating.
What if mating isn't random, like inbreeding?
Right.
Non -random mating, especially inbreeding, mating between relatives, does something subtle.
It changes the genotype frequencies specifically.
It increases the proportion of homozygotes.
But, and this is key, it does not change the overall allele frequencies in the population by itself.
So it shuffles the deck, increasing pairs, but doesn't remove any cards.
That's a good way to put it.
It rearranges alleles into genotypes, making those rare recessive alleles more likely to show up in homozygous form, but it doesn't make the alleles themselves more or less common overall.
We use the coefficient of inbreeding, F, to quantify this effect.
Okay, so we've got these forces, selection,
drift, migration, mutation -changing allele frequencies within populations,
microevolution.
But how do we get from that to, well, entirely new species,
macroevolution?
Yeah, this is where things get really fascinating.
The absolute key concept here is reproductive isolation.
Speciation is fundamentally about populations diverging genetically to the point where they just can't successfully interbreed anymore.
So they become separate gene pools.
Exactly.
And the barriers preventing interbreeding fall into two main camps.
Resugotic barriers act before fertilization.
Maybe they live in different places or have different mating seasons or rituals, or their reproductive structures just don't fit.
So no zygote even forms.
Correct.
Then you have post -zygotic barriers.
These happen after fertilization.
Maybe a hybrid zygote forms, but it doesn't develop properly, or the hybrid offspring is born but it's sterile, like a mule, or it has really low survival or fertility.
Either way, it's a genetic dead end.
Is there a good real -world example of this process?
Oh, the snapping shrimp in Panama are a fantastic case study.
About three million years ago, the isthmus of Panama rose, separating the Atlantic and Pacific Oceans, the split -inceptual shrimp populations.
And they went their separate ways, evolutionarily speaking.
They did.
Today, if you take pairs of the most closely related species from the Atlantic and Pacific sides and put them together in a lab, you find strong reproductive isolation.
Mostly prezygotic, they often just aren't interested in mating with the wrong kind.
But even if they do mate, post -zygotic barriers like hybrid inviability or sterility are common.
Speciation is nearly complete.
That took millions of years.
Does it always take that long?
Not necessarily.
Speciation can sometimes be remarkably fast.
Uh, look at the cichlid fish and Lake Apoyo in Nicaragua.
There's strong evidence, including from mitochondrial DNA, that one species, the arrow cichlid, evolved from another, the mitis cichlid, within that single lake in less than 10 ,000 years.
Wow, in the same lake.
How did they become isolated?
It seems likely driven by disruptive selection.
Possibly related to feeding niches.
Combined with assortative mating, they started preferring to mate with fish that look like themselves.
They developed reproductive isolation surprisingly quickly.
So one species form, we want to figure out how they're all related, right?
Build the tree of life.
That's phylogenetics.
Exactly.
Phylogenetic trees are hypotheses about evolutionary relationships.
They show patterns of common ancestry.
The branch points, or nodes,
represent the last common ancestor of the lineages splitting off.
And today, DNA sequencing is overwhelmingly the main tool.
How does that work, basically?
Well, you sequence genes or whole genomes from different species, align the sequences, and compare the differences.
Species that share a more recent common ancestor will generally have fewer differences in their DNA than species that diverged longer ago.
Computer algorithms use these patterns to build the most likely tree.
Has DNA sequencing settled any big evolutionary questions?
Oh, absolutely.
For a long time, there was debate about whether the coelacanth or the lungfish was the closest living relative of land vertebrates.
You know, tetrapods.
Both are fish, but kind of weird ones.
Right.
Living fossils.
Sorta.
Anyway, by comparing sequences from hundreds of genes across many vertebrates, phylogenetic analysis pretty definitively showed that the lungfish shares a more recent common ancestor with us land dwellers than the coelacanth does.
DNA settled that one.
And can we put dates on these splits in the tree?
We can estimate them using molecular clocks.
The idea is that some DNA changes, especially neutral ones, might accumulate at a relatively constant rate over time.
If you can calibrate that rate using fossils with known ages, you can then estimate divergence times for lineages where you don't have good fossils.
Like the human -chimp split.
Yeah.
Molecular clock estimates calibrated with fossils suggest that the human and chimpanzee lineages last shared a common ancestor somewhere around 7 to 10 million years ago.
OK.
Let's end with maybe the most exciting area right now.
Our own story.
Using ancient DNA.
Paleogenomics.
It's completely revolutionizing our understanding of recent human evolution.
Sequencing DNA from extinct hominins like Neanderthals and the less well -known Denisovans has been incredible.
Because it challenges that simple out -of -Africa and replacement idea.
Exactly.
The big finding is that when modern humans migrated out of Africa, they encountered these other hominin groups and they interbred.
It wasn't just replacement.
There was mixing.
And we carry that history in our genomes today.
We absolutely do.
People with an ancestry outside of Africa typically carry about, say, 2 % Neanderthal DNA on average.
2%.
Yeah.
And some populations, particularly in Melanesia, carry an even higher percentage, maybe 5 % or 6 % of Denisovan DNA.
So our genomes are actually mosaics, reflecting this complex history of migration, divergence, and then, well, renewed gene flow between groups that were pretty distinct, maybe even separate species by some definitions.
Hashtag tag outro.
Wow.
What a tour.
We went from the sheer amount of hidden genetic variation
through the elegant math of Hardy -Weinberg as a baseline.
To the actual forces like selection and drift that push populations away from that equilibrium, causing microevolution.
All the way to how reproductive isolation builds up, leading to new species, and how we use DNA, even ancient DNA, to reconstruct that vast evolutionary history.
It really shows how population genetics gives us conceptual and analytical tools, the math, the molecular data, to study evolution at every level, from single genes changing frequency to the grand patterns of lice diversification.
So thinking about that ancient DNA story, maybe the final thought for you listening is this.
As we uncover more about interbreeding between ancient human groups like Neanderthals, Denisovans, and modern humans, how does that force us to rethink what we even mean by a species, especially when talking about our own recent past?
Yeah, that's a really fundamental question these discoveries keep pushing us to confront.
Definitely something to ponder.
Indeed.
Well, thank you for joining us for this deep dive into the foundations of population and evolutionary genetics.
We hope you found it useful.
Catch you next time.