The evolution of the genetic code

I wonder how ID proponents will spin these results.

Re "The evolution of the genetic code

PvM posted Entry 2893 on February 10, 2007 04:15 PM.
Trackback URL: http://degas.fdisk.net/cgi-bin/mt/mt-tb.fcgi/2883

Scientists from the Weizmann Institute of Science report the exiting findings of Parallel Codes In Genes"

exiting findings? Or exciting?

Henry

I wonder how ID proponents will spin these results.

I wonder how ID proponents will spin these results.

You mean you haven't seen this yet?

http://www.uncommondescent.com/archives/2055

Is this yet another Biblical Code?

Don't these findings imply that beneficial mutations are even more unlikely than previously believed? Not only do gene mutations need to improve the protein design, but also the parallel regulatory code.

I fully expect Sal to start blathering about steganographic images concealed in alternative codes.

Oh dear. Over at UD, they seem to think that this finding of a "parallel code" is actually another gene, with talk about the probability of it having "evolved" multiple times...

However, if one looks at the research, what they are saying is that the genetic code is very efficient at coding information for folding, regulation, splicing etc in parallel to the genes that actually code for proteins... It's a feature of ALL DNA, and hence evolved way back at the start of life on the planet.

The researchers believe that the efficiency of the code to perform multiple tasks is a hallmark that the genetic code itself evolved early on in the history of life.

A couple of points:

1) If it's true that this capability to be highly efficient evolved, that implies that we have metagenes (genes that describe genes) stashed away somewhere in our DNA. Finding this stuff would give us a great insight into what LUCA was like

2) Genetic algorithms are indeed well-known for using the same part of a "genome" to perform multiple tasks in a "phenome". I worked with evolving circuits on PLA chips back in my undergrad days (a follow on from work done by Adrian Thomson), and we found for example that an evolved circuit would use a loop, both to generate an RF signal that critically affected the behaviour of another part of the chip, but also at the same time to introduce a delay into another signal. At least, we think that's what the circuit was doing :-) Anyway, the point being that seeing this type of parallel use is what I would consider to be the very hallmark of a genetic algorithm, as opposed to an intelligent designer. There are very few examples in human history where we have designed a part of a machine that performs two tasks at once. Can anyone think of an example? It's horrendously difficult to do - think on two levels of conflicting abstraction at once.

Maybe someone with a valid logon over at UD should point this out to the IDists... :-)

Wade,

The short answer is - no.

The long answer is that if the mutation doesn't obey the rules needed for the genetic code to be respected, the mutation simply dies - the cell where the mutation occured doesn't survive to pass the mutation on. Why doesn't this make things harder to evolve? Simply because the rate of these "bad" mutations is probably many orders of magnitude higher than the rate of mutation that we generally talk about.

The reason your confusion arises is because you have used two different meanings of the word "mutation" in your question. Most of the time, when we talk about "mutation", we are talking about viable mutations - where viable means that the cell where the mutation occured is still able to reproduce, even if ultimately the organism dies. This was the sense of the word that you used when talking about a "beneficial mutation".

The second meaning of mutation is any error in the copying process of a cell's DNA. Most of the time, these errors produce DNA that is not capable of reproducing. The mutation dies and disappears.

This mechanism can be more usefully thought of as an error correction mechanism. If an error occurs during cell reproduction that is so wrong that a viable offspring cell can't be produced, the offspring cell is simply never created - saving energy for the organism... There are other studies around that show that we do infact have a fairly high failure rate when copying DNA during cell reproduction, but with the parallel code acting as a kind of checksum, our organism can automatically remove these types of errors. It's very neat :-)

I can tell you what evolution deniers will say.

Evolution is based on RANDOM, ACCIDENTAL mutations culled by NATURAL SELECTION.

Everytime you find MORE COMPLEXITY the mathematical and REALISTIC CHANCE that what we have become is ACCIDENTAL continues to EVAPORATE!!!

It was already ludicrous when we hit our first basic understandings of genetics. Now that we are finding that the actual complexity are orders of magnitude above what we THOUGHT it was it is ABSOLUTELY LUDICROUS TO THINK IT HAS BEEN AN ACCIDENT THAT JUST HAPPENED WITHIN EVEN 4.5 BILLION YEARS!!!

There are more mutations required per year than you have years!!!

HAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHA

Yup, I LOVE to hear about the INCREASING COMPLEXITY that implies TRILLIONS of negative mutations that FAILED!!

Oh, and you can argue Evolution v ID all you want. I simply can't believe the STUP!D!TY of arguing so hard for a THEORY you DON'T have FACTS to support!!

Just because you think your THEORY is the only reasonable one DOES NOT MEAN IT IS WORTH A SECOND GUESS!!

Maybe it is time to be looking for a MECHANISM that is FEASIBLE in REALITY??

HAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHA

I still have difficulties to imagine how the genetic code could evolve since any change in it should be lethal.
Yet it did.

Are there reseach articles about the evolution of the genetic code? It seems that this one doesn't discuss it, just the optimality of the "end product".

Thanks.

I still have difficulties to imagine how the genetic code could evolve since any change in it should be lethal. Yet it did. Are there research articles about the evolution of the genetic code? It seems that this one doesn't discuss it, just the optimality of the "end product". Thanks.

jerry at UD: "What are the odds that such a sophisticated mechanism evolved in the short time before the Cambrian."

Uhhh, ouch. Considering that the first life appeared 4 billion years ago, and the Cambrian started 543 million years ago, "the short time before the Cambrian" is actually 86% of the entire history of life on earth. But then, no one every accused the IDists of knowing their information.

It's really hard to read the ignorance in some of those responses.

Are there reseach articles about the evolution of the genetic code?

— jeannot

Wow! The comment of KuhnKat demonstrates once more the sophisticated level of ID science.

I have to say that one is one of my all-time favourites.

...especially the "HAHAH.." part. Imagine a scientific argument where one side actually behaves like that.

The article is (apparently) available without subscription at http://www.genome.org/cgi/reprint/gr.5987307v1 in the preprint (not yet published) section of the journal.

I downloaded the report and read all 9 pages of it. After reading it, I haven't satisfied with the experiment result, because:

1. The genetic change that could illustrate the evolution theory must not only add information to organism's genome, it must also add new information to the biocosm. The experiment didn't show that.

2.Read this table of 'Significance of the genetic code in representing arbitrary sequences' carefully:

n-mer size P-value average P-value 20%
log-probabilities

5 0.110 0.054
6 0.097 0.045
7 0.083 0.028
8 0.049 0.031
9 0.043 0.010
12 0.028 0.004
15 0.016 0.004
18 0.012 0.006
20 0.026 0.006
22 0.021 0.004
25 0.029 0.009
(Source: Shalev Itzkovitz and Uri Alon, "The genetic code is nearly optimal for allowing additional information within protein-coding sequences", page. 5). The table showed that the probabilities of allowing additional information within protein-coding sequences is actually zero a.k.a nil. It means: Not a chance!

3. The report said that additional information within protein-coding sequences is useful. If it is, so, amazingly, the researchers had figured out of how amino acid could turn into protein, If they had, so, they were able to built a new kind of protein that never existed in the world before. If they were, so, they had ability to construct a mechanism of universal genetic code evolution. But unfortunately, the report didn't tell all about that. Why? Because this experiment was based on mutation. Israeli biophysicist Lee Spetner said: "Evolution cannot be built by accumulating mutations that only degrade specificity", (Dr. Lee Spetner, "Lee Spetner/Edward Max Dialogue: Continuing an exchange with Dr. Edward E. Max," 2001, http://www.trueorigin.org/spetner2.asp )

4. The title of that report is: the genetic code is NEARLY optimal for allowing additional information within protein-coding sequences. In another words: the genetic code is ALMOST optimal for allowing additional information within protein-coding sequences. It means that genetic code is not yet optimal for it. So, we can not yet accept this experiment as an evidence of genetic code evolution.

Inal my thoughts exactly. Funny how your comment is the most recent and last one ;)