Friday, February 03, 2006

11. Evolution: Mathematically Possible

Many people argue that evolution, notwithstanding all evidence found by science, simply doesn't make sense. How can a bacteria, they say, simply become a human, even if it has a billion years to do it? How did homo habilis suddenly decide to "get taller" one day and "evolve" into modern man?

I will not devote a great deal of time here to natural selection; it is clear that, if a mutation in an individual organism is a favorable one, that organism will live on and reproduce. The organisms I will use in this pseud0-proof are two HIV strains that I work with on a daily basis, HXB2 and 93TH253; more specifically, I will talk about a gene known as the "RT gene" (if you are some form of biologist, I apologize for the oversimplification). I chose these two strains solely because I know a lot about them and I can probably answer any questions; these strains are no more adequate in proving my point than any other two strains.

HXB2 comprises a major branch on the HIV "family tree" (a.k.a. the phylogenetic tree, for those of you who know a little bit about genomics). It was first found in France in 1983, and a great deal of humans is now infected with one of the many mutated variations of HXB2.

One of these mutated variations, 93TH253, was found in 1993 in Thailand. The RT gene of this strain has well over 90 mutations, which comprises over 10% of the gene. Therefore, let's be safe and say that IT IS POSSIBLE for 10% of a gene in a simple organism to mutate within ten years.

Now let's assume that in only 20% of cases (1 out of 5) will this 10% change in a gene result in a 5% increase in gene size (this is, again, a very safe estimate) while the other 80% of cases will either result in insignificant size changes or a decrease in size.

Therefore, in ten years, a gene has a 20% chance of increasing in size by 5%.
Take 20% of 5% (equals 1%).
So, in ten years, a gene may very plausibly increase in size by 1%.
So, in ten years, a gene's size may plausibly increase to 101% (1.01 times) its original size.
If this is true, in a mere thirty thousand years, a gene's size can increase to 9 trillion (9,000,000,000,000) times its original size. After thirty thousand more years, it will be 9 trillion times that amount (81,000,000,000,000,000,000,000,000).

The human genome is about 3 trillion bases long and took over a billion years to create.

The human genome is one-third the size and took over thirty thousand times as long to create as this theoretical genome.

30,000 / (1/3) = 90,000.

Therefore, by this interpretation, there is a 1/90,000 chance that these calculations are incorrect and that evolution is, as a matter of fact, impossible.


Remember that all of these calculations are approximations and are not 100% correct... the actual probability may be 1/60,000 and it may be 1/120,000.

1 comment:

Edwin said...

I agree with the feasibility of evolution.

I am a computer engineer not a biologist, but I think your extrapolations of increase in gene size are wrong.

Because,
1. The increase in gene size during evolution need not be linear. Effects on gene size due to natural selection is more complex. See this

2. Gene Size need not be a direct indicator of functionality and complexity in the organism.

Digression
Infinite monkeys on infinite typewriters can type out all possible DNAs for all organisms.

Given enough time N monkeys can do the same job with N typewriters if N is a large number.

A God can do much better than the monkeys.

If I were God, I would have let the monkeys do the dirty work and I would have put my hands and minds on the finest.

As any good programmer would tell you, extensive use of automated code generation is a well known strategy in software engineering. It saves a lot of work by creating redundant and less relevant code while we develop the sophisticated modules.