GENETIC LINKAGE EDITION!
The concept of genetic linkage is tied in with the concept of genetic recombination (i.e. crossover). During meiosis, homologous pairs of chromosomes trade DNA by crossing over each other and swapping. This is after replication; where you've got two X-shaped chromosomes in each pair.
![](http://pics.livejournal.com/commeantennas/pic/0000d9r2)
This probably does way more to confuse than it does to help you understand ....
Two genes are linked when they're so close together that, when crossover occurs, they crossover together more often than not. This means that the inheriting of gene A and gene B together, if they are linked, isn't random; the probability of inheriting these two traits changes. By finding the percentage of recombinant offspring in a cross between a heterozygote (het) and a homozygous recessive (hr), you can find the rate of recombinance. (Crossing an organism of unknown genotype with an hr organism is called a testcross, and it's really cool because you know the genotype of all the offspring since you know that half of the alleles are recessive. For example, if you cross a het organism for trait A and a hr for the same trait, you know that if the offspring exhibit a dominant trait, they must be het, because only one parent could contribute a dominant allele.)
HEY EXAMPLE TIME!
So you've got two fruit flies. The female is het and the male is hr for genes A, B, and C. (ABC/abc x abc/abc) The distribution of their offspring's genotypes are as follows:
98 ABC/abc
90 abc/abc
30 Abc/abc
30 ABc/abc
28 aBC/abc
26 abC/abc
3 AbC/abc
5 aBc/abc
BEFORE I GO ON, I want to point out how important that it is that in fruit fly testcrosses used to determine recombinance frequency the male be the hr one, because male fruit flies don't have crossover. So you'd never get recombinants if you tried it with the female being hr.
SO. The ones that look just like the parents (ABC/abc and abc/abc) are called parental. They are always ALWAYS the most numerous, because recombination frequency, when it hits .5, is considered a sign that the genes are unlinked. Either you have equal numbers across the board, or you have fewer recombinations than parentals.
The guys in the middle (the 30-30-28-26 guys) are the ones that have single crossover-- that is, crossover between A and B, or B and C, but not between all three. The guys at the bottom are the double crossovers-- crossover between all three genes-- and these, as you can see, are SUPER RARE.
So here's the formula for finding the percentage of recombination between two alleles:
#recom/total.
When you've got three genes like this, you need to include the double crossovers when looking at only one gene. So, looking at crossover between A and B, you've got 30 + 28 + 3 + 5 for the #recom (66) over all of the offspring (310). 66/310 = .2129. Doing this for the second group of crossovers (between B and C) you have 30 + 36 + 3 + 5 (64) over 310, which is .2065. The percentage of double recombinance is 8/310 = .0258.
These probabilities are the recombination frequencies, and you can use them to map out genetic distance between the linked genes. 1% recombination frequency equals 1 map unit (m.u.) or 1 centimorgan (cM). I usually call them cM, although fruit fly geneticists lean towards m.u. and since we're talking about fruit flies, let's use them.
The recombination frequency between genes A and B is .2129--> 21% --> 21 m.u. The recombination frequency between genes B and C is .2065 --> 21%--> 21 m.u. You don't really need the percentage of double recombinants, but I figured I'd just go all the way through. Because I'm thorough like that.
Now you can draw a super fun diagram of the chromosome with the locations of the genes!
![](http://pics.livejournal.com/commeantennas/pic/0000eewa)
I'll try to find some more resources on this so you can get a possibly clearer explanation.... Although what I've found so far doesn't make as much sense as this does.
Next time: Lod scores!