All language subtitles for 87031x_PR_DNA_Markers_02_Seq_Location_Polymorphism_-en

PETER REDDIEN: Now you and I, all individuals in this room

will vary roughly about one nucleotide for every 1,000

nucleotides.

So humans are very, very similar to one another

at the DNA sequence level.

One in every 1,000 nucleotides, on average, are different.

But nonetheless, we have a 3.2 billion nucleotide genome.

and there could be heterozygosity and so on.

So there's roughly 4 to 5 million sequence differences

between to any two individuals.

So that's a lot.

So if you do the sequencing, you've

got 4 to 5 million differences to sort through.

So it's literally a needle in a haystack to try to find this.

So how in the world do we do it?

There's plenty of human disease genes that have been found.

You could take model organisms, like our paralyzed fly,

and you can find the gene and study it,

try to do things with it.

So how do we do it?

Sequence lots of individuals.

Well, so all the individuals are going

to have a lot of variants, and if there's some clonal origin,

let's say, of the variant like one paralyzed fly, or one

mutated individual, then all of those variants

will be present in all the descendants,

depending upon the nature of crosses that have happened.

But you could start working with that,

the nature of crosses that have happened, to think about this.

But if you sequence a lot of clonal individuals,

you just find the same polymorphisms over and over

again.

But there is something to what you're getting at.

Any other suggestions?

So the suggestion is look at how far apart things are.

But what things are you looking at

to see how far apart they are?

You're kind of on the right track.

OK, so look at other things that are

different in that individual that

might be going with that trait?

STUDENT: Yeah.

PETER REDDIEN: OK.

So that's good, and in general, that's

the approach taken in a lot of human genetics, which

we'll get into.

Did you have a--

So the suggestion was, if the location of another gene,

can you use that?

OK, that's great.

So what you could do, so you remember the genetic mapping

with yellow and vermillion and stuff, miniature wings on the X

chromosome.

Let's say you find, oh, I'm one centimorgan away

from one of these, and the other one

has been discovered from other work.

Well, then I might roughly know the area

of the genome this is in, because you've

placed some genetic marker onto a physical map,

a reference sequence.

And I'm in between these known--

from the genetic crosses, these known genetic markers.

And I know their position on the physical map,

so I know I must be in this region of the physical map.

So that's good.

That can start narrowing it down.

Great.

So I think we're getting on the right track.

Let's now dive in to trying to do this in as efficient a way

as we can think of.