Monday, September 17, 2012

Finnegans Genome

I had a little fun writing yesterday’s post. In today’s post I’m going to look at two other popular arguments against evolution. The first is that there is never any new genetic information produced, and the second is that all mutations are harmful.

These two assertions are often combined, so it is convenient to tackle them together in one post.

Natural selection merely rearranges genetic information supplied by God at the time of creation.
All mutations have proved to be harmful.

Yesterday’s post was crafted to show that probabilities are governed by rules. This is a central concept for the theory of evolution. Depending on the pathway to a particular sequence the likelihood of that sequence developing can be dramatically reduced. I gave one example of a sequence that might take billions of years to be realized developing in a quarter of an hour. The premise of yesterday’s post was independent of what the sequence would be. Today’s questions deal with the sequence.

One of the interesting things about these two questions is that they are based on incompatible premises. Mutations are essentially new information. So if mutations are at all (harmful or otherwise) there must be new information introduced into the genome.

There must be some other meaning for "information" or "new"in order for both of these questions to make sense at the same time.

I sometimes answer this question with a description of a simple experiment.

1) Grow a large batch of bacteria starting with a single bacterial cell that is susceptible to a particular antibiotic. The particular antibiotic binds to a particular transfer RNA molecule in the bacteria.  Sequence the tRNA gene.

2) Add more medium and the antibiotic to the culture. Without the antibiotic the number of bacteria would increase, but most of them are killed by the antibiotic. Some do grow, and eventually you have a large batch of resistant bacteria.  Sequence the tRNA gene.

3)  Notice that the two tRNA genes are different. The difference is new information introduced by a beneficial mutation.

This is such a terribly simple experiment that variations of it are being conducted right now in beginning microbiology labs all over the world. It takes about three days to perform. I could substitute any number of selective traits to generate the same result.

Mutations occur, they are the result of new information being incorporated into a genome, and through natural selection beneficial mutations are identified.

I know what you are probably thinking: “If it were that simple these questions would no longer be asked”. The reason why such an elegant experiment cannot dispel these questions is that the creationist is often using definitions of information and new that you may be unfamiliar with.

To be fair the concept of “information” is exciting and somewhat fluid. What do you need in order to really mean something? How can you tell if one thing means more than another? And perhaps most important:
What do you really mean by this?

There are objective measures of potential information content in a sequence. They can be applied to any sequence; even sequences of letters that make up works of literature. Two books do not necessarily have the same information if they use the same alphabet. Two poems that use the same number of each letter may be quite different. However, it becomes more likely that two works contain different information if they differ in word count, letter usage, or any number of content-independent measurements.

Meaning is more subjective. Two people reading Finnegans Wake might think it means two different things; for that matter one person reading Finnegans Wake might think it means two different things, and under the right circumstances might write a well-received dissertation contrasting them.

Subjectively the information in a sequence makes sense in context. Finnegans Wake, for instance, might make no sense in my hands; a Russian translation of it doubly so. This situation might appear to exert a stabilizing influence on the information content of a sequence by applying interpretation pressure from each observer, but it actually does the opposite. In the natural world the information in genomes is interpreted in a huge array of contexts, and if meaning is tied to context there is a huge amount of continually new information in the genome with no changes in the underlying sequence.

It is through natural selection that the meaning is most conclusively interpreted.

If the idea of information can be subjective the idea of harm must be positively arbitrary. 

I remember being told the parable of the antlers more than once.  Out in the open field the bucks with the largest antlers were the most successful at passing along their genomes, but when the hunters came the small antlered bucks disappeared into the densest brush while their big-antlered cousins got hung up.  I think there is an allusion to the story of Absalom in that parable. 

The tome that is Finnegans Wake can also serve to illustrate this point.  A graduate student that discovers a committee that shares their obsession with Finnegans Wake might think think that graduate school is a breeze, but once they graduate they will be eating cat food while decrying the lack of academic jobs. 

No comments: