Really? What two sequences are those Bill? Quote him. So there’s few green bricks, and lots of brown bricks, and only shots in the green bricks. That’s his analogy.
Where are the 2 sequences, Bill?
1 Like
Why do you reject negative selection of deleterious mutations as a mechanism?
You can’t measure functional information by comparing two proteins. Comparing two proteins can not tell you all of the possible protein sequences with that specific function.
Broken records aren’t science.
1 Like
I don’t think we are being productive here.
I think I am. I think you’re now trying to get out of admitting you were blustering. Let me give you that out: Admit it.
2 Likes
Model it creating complex adaptions.
Arbitrary. I don’t think this conversation is productive.
The model starts with a complex adaptation, the protein under question. Starting with a functional protein, can you please tell us why you reject the model where deleterious mutations are selected against through natural selection?
1 Like
Why are the deleterious mutations being selected against?
Individuals who carry deleterious mutations tend to have fewer grandchildren which reduces the number of individuals carrying that mutation in the population. It is basic natural selection.
1 Like
It could also be that certain deleterious mutations are stopping embryonic development. Which I would suspect is the case with proteins like actin that are so highly preserved.
Do you now understand why natural selection would cause proteins in different lineages to have fewer differences than we would expect from neutral drift? Do you understand how natural selection drives sequence conservation?
2 Likes
I understanding this is a possible explanation. When preservation is extreme this does not look like a viable explanation.
Do you make bare assertions with absolutely zero explanation just to troll us?
5 Likes
Hi Bill,
I’d like to provide a couple of examples from other domains to illustrate key information theory concepts that are being discussed here. I think you’ll understand better why the biologists are kinda worked up over the issue, and you’ll be able to have a more productive conversation with them. Or maybe you’ll just want to cogitate for a while on these illustrations. Whatever works best for you.
Information Defined
Shannon defined information as one or more bits (in the computational sense) of message that reduce uncertainty about the message being transmitted.
Say for example that you are receiving a message by Morse Code. The incoming message arrives a letter at a time like this:
I-W-A-N-T-T-O-A-S-K-Y-O-U-A-Q
There is no point in sending the U-E-S-T-I-O-N, as they would add no information in the Shannon sense. You already understand from the context what the next 7 letters of the message will be. Sending those 7 letters would not provide you with more information.
Measuring Shannon Information
In fact, the best way to measure Shannon information is by the number of bits in the message after it has been maximally compressed. A Word document that’s 10kb in size does not contain 10kb of information because you can compress it with Zip down to 2 - 3kb.
But that’s not the only consideration here. You can always find ways to make a message more terse, and thus contain fewer bits of information. This is why young folks can send text messages far more quickly than adults. In the time that it takes me to type “lots of laughs, I’m rolling on the floor laughing,” my kids can zip 3 or 4 messages back and forth because they’ll type “LOL ROTFL.” You also see this in the difference between pedestrian and precise prose. The poorly written stuff meanders on paragraph after paragraph without saying much, while the concise stuff says a lot more with a lot fewer words.
So just because a sequence of words in a text file consumes 10kb of space on disk does not mean that it contains 10kb of information. You could shrink it by writing it more concisely. Then you could shrink that concise version by a compression algorithm. Even then you have no guarantee that you have a 100% efficient information format; maybe you could rewrite the message even more concisely, or maybe you could use a better compression algorithm.
Kolmogorov Complexity
The Kolmogorov Complexity (KC) of an object is the length of the shortest program that can be written in a given programming language to produce the object as output. Lincoln’s Gettysburg Address is quite compact, so pretty much the only way you could display it to console in Python would be to write every single character of the speech into a string and call print():
address = """Four score and seven years ago....""" # etc.
print(address)
The program that prints a song by Yes could be much shorter than the song itself, however:
refrain = """I've seen all good people turn their heads each day,
So satisfied I'm on my way"""
for i in range(11):
print(refrain)
One of the interesting qualities of KC is that you cannot prove you have ever found its lower bound. Maybe there’s a way to write the program more concisely! How would you prove there is not?
So the KC of the Gettysburg address is probably very close to its Shannon information, whereas the KC of “I’ve Seen All the Good People” by Yes is at least an order of magnitude smaller than its Shannon information (assuming the listener was surprised at each repetition).
Application to DNA Sequences
When the biologists say that sequence conservation does not equate to functional information, they are applying the principles of information theory. There could very well be a way to specify the same function with a far smaller sequence of nucleotides, just as you could specify the same text message with fewer keystrokes (LOL!). Switching over to the Kolmogorov complexity measurement, there could very well be a way to specify the same function with a far smaller DNA sequence. Thus the length of a DNA sequence is not the same thing as its functional information.
Biologists, I have done my best to understand your argument and recast it in explicitly information theory terms. I hope I have not done any violence to your ideas. If I have, please feel free to correct whatever needs correction.
Bill, I hope you have found this helpful.
Best,
Chris
2 Likes
This is why we need a (facepalm) emoji.
2 Likes
It exists! 
1 Like
But you just supplied one. You literally just gave a negative selection explanation for extreme conservation. Your explanation was that mutations in the conserved proteins are “stopping embryonic development”. So a mutation that stops embryonic development is strongly deleterous, right? So negative selection is keeping the protein conserved.
3 Likes
Thanks for the thoughtful reply. This is something that could be explored. The data so far does not seem to support this over all proteins but could be a possibility.
The way Gpuccio deals with this is the there is large room for error in his calculations.
I think you stated this idea well. I think debate here is healthy as long as it does not contain a lot of arbitrary assertions, question begging and straw man arguments. I am simply advocating that design be considered as an alternative explanation for what we see inside the cell.
So, Bill, how much money do you want to bet that your suspicion is correct–that any mutation in the beta-actin (there are three, not one) gene in humans is embryonic lethal?
Why do you shy away from the scientific language of hypotheses and predictions and experiments for the safety of arguments and suspicions?
I really don’t understand your point here.
Gpuccio uses the analogy of the wall made of brown and green bricks to show that if the function of an object is an objective property of the object, we are not at all committing the TSS fallacy even if we recognize the function post hoc. And since the functions of many proteins are objective properties of these proteins, no TSS fallacy at all is committed by ID proponents when they recognize the function of a protein post hoc. Period.
That’s not the only thing he is doing with that analogy. He also says the wall is being shot at, and that:
We also observe that all the 100 shots have hit green bricks. No brown brick has been hit.
Then we infer aiming.
The contention that some of the bricks(proteins) have functions is not what I am objecting to.
Can you explain to me what are the shots? Green bricks are functional proteins, right? Brown bricks are nonfunctional proteins, right? So what are the shots and what does that have to do with evolution?
I fail to see how this analogy mirrors anything in real biology. Any attempt to make it reflect something biologically real would make it obvious how you’d either end up making the sharpshooter fallacy, or simply assume the claim that functional proteins are so rare as to be impossible to hit with the available number of un-aimed shots.
2 Likes