I had a entire discussion about Irreducible Complexity and used Richard Lenksi's E. Coli experiment has evidence that 1) new information does arise through mutations and that 2) Irreducible Complex structures do form through mutation. It's rather sad that there is an abundance of evidence and it is ignored.

Yes, Lenski's experiment is a great testament to evolution in action.


"But that's 'micro-evolution'!" .......

So according to that comic, we're all doomed ?

Yeah now I get why they need their religion to keep high moral

JoeB wrote:Yes, Lenski's experiment is a great testament to evolution in action.

"But that's 'micro-evolution'!" .......

And it's not going to turn a crocodile into a duck, is it?

I was wondering by the title if something so stupid had to make all the users on this site to post a facepalm pic, could it potential destroy internet?

probably not, but that would :

[youtube]http://www.youtube.com/watch?v=gA9W9NHgUko[/youtube]

aban57 wrote:probably not, but that would :

http://www.youtube.com/watch?v=gA9W9NHgUko

please no flash photography you can damage the internet. LoL

I want to watch all the IT crowds now.

JoeB wrote:It seems their entire argument comes down to their idea that no new (genetic) information is created through evolution. Is this correct? (I assume not, just checking as I'm no biologist).

This depends on what definition of information you apply to what part of the process. There are several mathematical definitions of information - and without stating which one one is talking about it´s hard to tell what it could mean. These definitions generally describe some feature of a random variable and evolutionary biology is generally described using stochastic processes, i.e. sequences of random variables. Now you can take any of these processes and see what happens to any of your information definitions through the sequence. And sure, in some cases you find information tends to get lost. In some cases you find the reverse. But the basic question is whether the stochastic processes accurately describe what happens and that´s been tested for decades (and revisions to make our descriptions more accurate have been made as well).

I´m going to pick one measure of information because it´s a rather cool example: Shannon entropy is a measure applicable to discrete random variables and for any discrete trait we can look at histogram space - that is, we tabulate frequencies for all possible states of the trait in the population. Any particular state then is a probability distribution and our random variable is what state we find when we pick a member of the population with a uniform distribution (i.e. any individual is equally likely). As it turns out for some traits the Shannon entropy over time keeps increasing towards a maximum value. For other traits the Shannon entropy goes to 0 (and since it´s non-negative, that is the minimum value it can take). There´s basically only these two ways it can go and a discrete trait that follows the second route is precisely what we call a gene (in population genetics).

In a less formal sense we can see that the maximum value is reached when every member of the population has a different state and the minimum when they all have the same state. For instance entire genomes follow the first dynamics - even identical twins tend to have mutations that make them slightly different from one another. On the other hand almost everybody has the same amino acid sequence for hemoglobin, thus this sequences is a gene. We can also note that mutations lead to new variation, while selection and drift (population resampling) remove it. So what our definition of a gene by shannon entropy does is that is looks at how fast new variation arises and how fast it is removed and based on these rates we call those traits genes for which variation is removed faster than it appears.