Original subtitles

PETER REDDIEN: OK.

So now, let's come back to our cross,

with this concept in mind, and think about what

happens to the next generation.

We do this cross with the sibling.

And we need to predict, then, what

happens in this F2 generation, with respect to the genotypes.

Now, how do we predict what happens?

Now, to do this, we're going to use Mendel's first law,

the law of equal segregation.

Equal segregation refers to this principle from Mendel

that allele pairs separate equally

during gamete formation, and then

unite randomly at fertilization.

So when this individual is producing gametes--

if it's a male, sperm--

then it's sort of random sampling of these two

alleles of the gene.

So, it's 50% chance at a given gamete--

sperm cell, say-- gets this allele, 50% chance

it gets that allele.

Same thing for gamete formation in the female,

and then when they unite, it doesn't really

matter what the nature of the allele

is for the probability of uniting.

OK, so let's just write this out a little bit more.

So our F1 gametes, we can look at the probability of a given

gamete.

We can say, what's the probability of a given gamete

here, having the TS allele?

And, what's the probability of having the wild-type allele?

Probability, if you sample a random gamete,

would be 0.5 for each.

So, these are the expected frequencies

of these gamete classes would be being produced.

OK.

So, we're going to make these gametes

and then unite them, randomly.

So, let's think about the combinations that could emerge.

Let's say this is the male, and this is the female.

We could get a para(TS) allele from the sperm,

and a para(TS) allele from the egg.

That's one possibility.

That's one possible F2 genotype.

We could get a para(TS) allele from the sperm,

and a para wild-type allele from the egg.

We could get the para wild-type allele

from the sperm, para(TS) from the egg.

And finally, we could get two para wild-type alleles.

Those are all the possible combinations

of what could happen.

And then, how do we determine our expected frequencies

of these classes?

Well, using this principle of equal segregation

where they're randomly uniting, then we

would expect these classes to emerge at equal frequency.

One way of looking at this would be

to say, what is the probability of event A happening,

and the probability of event B happening.

So, if what has happened is, you inherited

a para(TS) allele from the father and a para(TS) allele

from the mother, then this would be event A and event B.

And if event A and B are independent events, then

the probability of getting event A and B

is equal to the probability of event A times

the probability of event B.

This is the product rule of probability.

So it's like, if you're flipping a coin,

you've got a 50% chance of being heads or tails.

You say, what are the odds of getting

a heads on the first slip and a heads on the second flip?

0.5 chance of getting a heads.

On the second flip, 0.5 chance of getting a heads.

So it'd be 0.5 times 0.5.

OK.

Similarly, here, the probability of getting para(TS),

para(TS) is 0.5 times 0.5, equals 0.25.

Same for all of these classes, 0.25.

OK.

Now, you will notice that these two classes are essentially

the same genotype.

OK?

para(TS), para wild-type, one copy of each allele.

And so, we see that these genotypes

exist at relative frequencies of 1 to 2 to 1.

So we'll have a 1 to 2 to 1 ratio

of these different genotype classes expected.

OK.

Now, if we think about the phenotypes,

we see that all of these--

what would the phenotype of this first class be?

Anybody?

STUDENT: Paralyzed.

PETER REDDIEN: Paralyzed.

This second class?

Not paralyzed.

This class, not paralyzed.

OK, so all three of these are not paralyzed.

OK.

So, our expected phenotype ratios then are 1 to 3.

OK?

These three are expected to have the same phenotype.

And in our example data, we got a 1 to 3 ratio.

All right.

So, that's sort of what we would expect

with this hypothesis one.

And you'll see that this phenotype class

of not paralyzed by this model, by this hypothesis,

is sort of--

this 1 to 3 ratio-- is sort of masking a true 1 to 2 to 1

ratio of genotypes, where these not paralyzed flies could

fall into two categories.

Can anyone describe, or propose an experiment

that you could do that would distinguish between these two

genotype classes?

Keep in mind, we have two breeding strains.

We talked about this last time.

So yeah, you could breed the not paralyzed flies

to a paralyzed strain.

And what you see is, they'd break down into two categories.

I'll let you work this out for yourself after

lecture some time.

One class would give only not paralyzed flies.

The other class would get 50% of the progeny to be paralyzed.

OK.

OK, good.

So that's hypothesis one.