« Sock Puppetry/Happy Esperanto Day!Cthulhu Christmas »

Crossovers, Mutations and Assortment


  07:37:00 pm, by Nimble   , 1178 words  
Categories: Thoughts, Science

Crossovers, Mutations and Assortment

Something someone wrote today reminded me that in the discussion of biology and evolutionary theory, everyone looks at mutations, mutations, mutations, but they miss out the extremely powerful silent partners, chromosomal crossover and its most excellent cousin, independent assortment.

Crossover and independent assortment is the team that gives your children that extra bit of randomness, and by randomness, I mean an unpredictable mix of the genes that were contributed to you by your mother and father.

Otherwise, you would have up to four possible types of children, and by types of children, I mean exact types. Your second child would have a one in four chance of being as identical as an identical twin to the first child. If you had five children, at least one of them would be a repeat.

Independent assortment gives a powerful tool for mixing genes up.

Humans have 23 chromosomes, which are actually 46 tied-together chromatids. You get one chromatid from each parent.

Rough diagram of a single chromosome:

chromosome diagram

All the randomization happens before the sperm and egg get together.

In independent assortment, the chromatid from each chromosome gets selected at random. The biological equivalent of a coin toss. This gives a much, much higher assortment of possible children. Each sperm or each egg can be one of approximately 8.4 million combinations (coin flip times 23 chromosomes = 2^23 = 8,388,608). Combine one of each, and you get about 70 quadrillion (70 x 1013) possibilities.

...but wait, there's more...

There are also crossovers when DNA is dividing in the first steps to becoming sperm or eggs. This happens when there are double sets of each chromosome. Chromosomes replicate themselves just before most divisions, but crossovers only happen early on when cells are on their way to becoming sperm or eggs, in a phase called Prophase I.

Often, crossovers will happen several times in the course of this process. They line up and form what is called a Holliday Structure (despite the time of year, no, that's not a typo :) ). How the Holliday Structure is split affects the combination.

Holliday structure

This whole process not only further ratchets up the combinations, extraordinarily varied as they are, but it also allows genes on the same chromosomes to move in with new neighbours. This means that if you have a gene for something good (say, burning more calories in our modern environment), you can split it away from a gene for something bad (say, anemia). Now this also means that you can combine two bad things previously separated between your chromosomes, but experiments on the whole are good, when it comes to overall survival of the species.

Now let's actually go back to mutations for a second.

Most mutations do not actually do much.

Every three 'letters' in your DNA that actually get read are transcribed into proteins. Each set of three letters is a codon, and each letter can be one of four chemicals: Adenine, Thymine, Guanine, and Cytosine, usually abbreviated to A, T, G and C (and forming letters for the movie Gattaca)

This gives 64 combinations (3 letters in a codon to the power of 4 possible chemicals), but there are only 20 amino acids. There are six different "spellings" that code for the amino acid Leucine, for example. So some mutations may literally have no effect (though I imagine there might be slight speed differences between how fast the different spellings can operate)

However, even when there is a change in spelling (a "point mutation"), the effect may not be huge. Many proteins get their shape in part because some amino acids are hydrophilic, that is, they are attracted to water, and others are hydrophobic, that is, repelled by water.

(Soap, for example, works because one end is hydrophilic, and is pulled to the water, and the other is hydrophobic, and tends to surround dirt and grease)

The genetic code is even more redundant than it seems, because similar spellings usually get transcribed into amino acids with similar properties, but with small changes. This can slightly change the shape of a protein, or make it more or less able to dissolve in water, or what have you.

Here, there is a most excellent diagram of the amino acids (though it uses 1-letter codes, which you can find on Wikipedia). Slight changes in spelling typically do not move very "far" in this diagram.

Now I come back to crossovers and independent assortment...

Remember that there are almost always (with the notable exception of the X and Y genes) two equivalent genes in every cell, one from the mother, one from the father. Sometimes the action of one gene dominates, other times it contributes equally with its partner. Often, the gene it is partnered with can make a significant different in the effect the gene has.

Independent assortment lets nature try out the new mutation, particularly in small populations, where the new mutation is a much higher percentage of the gene pool. Crossovers let the gene be tried out in combination with other neighbouring genes.

Without shuffling that mutation into other combinations, good mutations can get lost if its neighbors are bad. This is the magic that sexual reproduction provides, and it is powerful stuff indeed, especially for bigger creatures that don't have the luxury of dividing into populations of trillions.

(Remember, too, that some mutations are "good" simply from the point of view of being "different" - bacteria, viruses and parasites can be rebuffed by unfamiliar proteins)

That said, new mutations are a fairly small part of the picture. Crossovers and independent assortment are never done playing with old mutations. Take blood groups: they are a very old line of mutations indeed, but the A, B and O groups are still being mixed in human populations.

So, when you hear about evolution consisting of mutations and natural selection, don't forget to pay a moment's respect to the big brothers of sexual reproduction, Crossover and Independent Assortment, that are responsible for most of the change we see.

Other odd information: when I was researching a few things, a question came to mind: does the intactness of the DNA of sperm affect how they swim, etc.? I couldn't find a direct answer, but I did find an interesting study where they irradiated cow sperm to see what happened. Apparently, irradiated sperm show no particular signs of problems, and in fact can fertilize an egg, but will stop dividing after two or three divisons:

he results show that sperm DNA damage does not impair fertilization of the oocyte or completion of the first 2-3 cleavages, but blocks blastocyst formation by inducing apoptosis [planned cell death].

I also found out that the crossover sites do not typically occur randomly, but have strong hot spots every 30,000 letters or so. That makes a lot more sense to me than mere random sites - DNA usually seems to have something special for certain functions. For example, there are CGCGCG... areas in DNA that seem to be almost like sacrificial anodes on boats (a block of a certain metal that does all the rusting, keeping your boat itself free of rust), electrically pulling free radical damage away from active genes.

No feedback yet