davidH

Yeah I understand so don't worry about it.

(when I made that statement about the introns I made a mistake! I did mean to say that except in the rare case a mutation in an intron wouldn't wreck the structure - sorry about that.)

But going back to my last post on page 3.

1. The occurrence of mutations is extremely rare, between 1 and 30 mutations per million gametes.

2. The occurence of a mutation that actually benefits the organism is even rarer again, with most mutations being fatal to the organism.

3.Duplication of the genes are also fatal except in the rare case.

4. The mutation has to occur in the exon, but possibly half the DNA or more is made up of introns.

5. 2% of the exons (or all the DNA - not sure on this) actually code for proteins.

Is anyone here a mathematician? What are the odds of a good mutation occurring when all this is considered.

Then what are the odds of mutations continually occurring until we humans have been formed?

Maybe I am wrong but for 1 mutation to occur that wouldn't kill the organism but remain useless or cause a benefit the odds must be over a billion: 1
Come to think of it they must be even more.
As the complexity of the organism increases then these odds would increase again right?

Also, at a certain stage of life the mutations would have to occur in the gametes? ie. sex cells to be carried on? So mutations on the surface ie not the sex cells, wouldn't benefit the offspring of the organism right?

If that's true then the odds increase even more.

I'd be interested on your views about this.

DNAunion

DNAunion: DavidH, you seem to still be shaky on introns vs. exons.

Say you have an imaginary gene with the following DNA sequence (note I just randomly typed some As, Cs, Ts, and Gs):


DNA: ACTGGGATCTAGATCGATAGCTATATATTTTCGCAATCGAT

It would be transcribed into a primary RNA transcript that is complementary to the DNA sequence from which the information was obtained. Complementary bases are those that pair together naturally in the strongest manner: G pairs with C and vice versa, and A pairs with T and vice versa in DNA and pairs with U and vice versa in RNA.

DNA: ACTGGGATCTAGATCGATAGCTATATATTTTCGCAATCGAT
RNA: UGACCCUAGAUCUAGCUAUCGAUAUAUAAAAGCGUUAGCUA

So I just replaced every DNA base with its RNA complement, which mimicks what nature would do.

One could look at the imaginary RNA sequence and determine what amino acids are expected to be in the protein produced by the gene. But through investigation it is found that the final protein is missing a large section and it is then found that the RNA used to produce the protein is:

RNA2:UGACCCUAGAUCUAAAAGCGUUAGCUA

Comparing the final mRNA used in protein synthesis to the originally transcribed pre-mRNA, we can find the segment that was omitted:

RNA1:UGACCCUAGAUCUAGCUAUCGAUAUAUAAAAGCGUUAGCUA
RNA2:UGACCCUAGAUCU--------------AAAAGCGUUAGCUA

Now we can figure out what are the exons and introns.

RNA1:UGACCCUAGAUCU[AGCUAUCGAUAUAU]AAAAGCGUUAGCUA

During processing of the pre-mRNA, the middle sequence (in brackets) is eliminated and the two ends are joined together.

The parts that make into the final mRNA - the two ends here - that are then actually translated into a protein are the sequences that are **EX**pressed. They are the **EX**ons.

The segment that exists between the exons and is removed is an **INT**ervening sequence. It is an **INT**ron.

[ February 03, 2002: Message edited by: DNAunion ]</p>

DNAunion

DNAunion: The first clause is correct: the second is not.

Most mutations are deleterious (that is, disadvantageous) to the possessing organism. But being "bad" is not the same as being lethal.

In fact, many argue that most mutations are neutral: neither advantageous nor deleterious.

So to please all parties, I guess a better way to state the issues with the effects of mutations is something like:

DNAunion

DNAunion: Gene duplications are not usually fatal. In fact, doubling of a given gene does not typically have any negative impact (unless, for example, it leads to overproduction of a protein with the excess number of copies of that protein upsetting the "balance" in the cell).

The general idea is that gene duplication creates an extra copy of the original gene. There are then two copies of which only one is needed. So one of the two copies can incur a mutation without it negatively impacting the cell. In fact, one copy (original or duplicate) can accumulate several mutations without a negative impact because the other copy of the gene (say the original one) is still being expressed and producing the needed protein.

So you have a DNA sequence which already produces functional protein domains (nature isn't starting the search for new information from scratch) and then mutations can occur in it without being deleterious. It is possible that after a series of mutations has accumulated that the copy will produce a protein with a new or slightly modified function, thereby proving to be beneficial to the possessor, and thus, retained by natural selection (assuming 100% efficiency of NS, which is not the case).

Although that is possible, it is not the usual result. Typically, a duplicate just accumulates more and more mutations and never gains any new or modified function, but having lost its original function: it turns into useless "junk DNA".

[ February 03, 2002: Message edited by: DNAunion ]</p>

DNAunion

DNAunion: In general, all the cells in your body (and the bodies of other animals that reproduce sexually) can be classified as being either somatic cells (body cells) or sex cells (gametes).

Body cells are those that make up things like skin, heart, eyes, legs, etc. They are what most of us think of when we think of a cell of an animal.

Sex cells are the eggs and sperm. They are the ones that fuse together, via sex, to produce offspring.

A cell in a male's eye does not fuse with a female's egg to produce offspring. A cell in a female's back does not fuse with a male's sperm to produce offspring. Somatic cells are not the cells that particpate in fertilization: only the sex cells do.

So any mutations that occur in a somatic cell can spread through that single organism by mitotic cell divisions. If a mutation occurs in gene X in a single cell in my eye, that cell can divide to produce to daughter cells that both have that mutation, and each of those can also divide with all four cells having the mutation, and so on, and so on. But this "eye mutation" will not be passed on to any of my offspring.

For a mutation to make into the next generation, it must occur in a gamete, and then that gamete must be one that participates in fertilization (a human male produces "gazillions" of sperm of which typically only 2 ever produce offspring) with the resulting offspring surviving to reproduce.

DNAunion

DNAunion: This response of mine will be rushed (going to meet the family for brunch).

A general rule of thumb is that changes are tolerated more readily in less complex systems. As systems become more complex and interdependencies arise (either between their own parts, or between them and other systems), tolerance to changes decreases. In complex systems, there are more constraints on what will work, so the number of changes that meet all of the criteria drops: consequently, it is harder - but not impossible - to hit upon a successful change by random processes.

theyeti

Yes, this is where pseudogenes typically come from. But the current in vouge theory on gene duplication is what's called the "subfunctionalization" model. The idea is that the gene (or rather the protein that it codes for) already has multiple functions prior to duplication. As is typical with many enzymes, they do one thing very well and a whole lot of things somewhat poorly. After duplication, one of the new genes specializes in one of the "poor" jobs and the other keeps doing what it did before. The evidence is showing that duplicated genes get preserved in a functional state more freqeuntly than previously thought.

See this recent article for some good info:

T. Massingham, L. J. Davies, P. Liò Analysing gene function after duplication. <a href="http://www3.interscience.wiley.com/cgi-bin/issuetoc?Type=DD&ID=85513053" target="_blank">BioEssays</a> Volume 23, Issue 10, 2001. Pages: 873-876.

theyeti

lpetrich

As to junk DNA, I have come across or discovered an interesting hypothesis: that it is really sacrificial, that it soaks up mutagens so those mutagens will have much less likelyhood of reacting with the important bits of DNA.

This hypothesis only requires that junk DNA be present -- it implies no constraints on junk-DNA sequences, which is consistent with its observed junk nature.

Has anyone else ever seen that hypothesis?

theyeti

No, and I don't know that it's viable. Most mutagens, like radiation, will work on all DNA equally regardless of the presence of other DNA. Having more DNA just means that you have more mutations; you still have the same number per gene. Although I can see how this idea might work with some chemical mutatgens (only those that get "used up", like nucleoside analogues), I seriously doubt that this is significant enough to warrant the presence of "junk" DNA. This hypothesis also fails to explain the general absence of "junk" DNA in bacteria, which are more frequently exposed to mutagens.

theyeti

theyeti

In humans (and other apes) I think it's 24%. Exons by contrast make up only 2-3%. In faster reproducing organisms, the intron percentage is much smaller.

Mutations aren't really rare, but beneficial ones are. Keep in mind that "beneficial" only has meaning with respect to the environment. If an organism finds itself in a new and unique environment, then the odds of a beneficial mutation are much greater, because the organism is not well adapted. Beneficial muations are rare simply because most organisms are very well adapted to their environments; it's an expected consequece of natural selection.

Pointing out that beneficial mutations are rare helps explain what we already know from Earth history. Evolution is slow. Human beings didn't pop up overnight; there were hundreds of millions of years if evolution that preceeded us. I don't know that any sort of "design" theory can adequately explain that fact.

theyeti