Functional Information (again)

The equation describes functional complexity.

You are randomly mutating through a sequence. Beneficial gene mutations are rare and deleterious mutations are most likely inside exons. With time you will move away from optimal function based on the ratio of beneficial to deleterious mutations unless the mutations are being eliminated by purifying selection.

No, it defines functional information. The way you use language prevents coherent discussion. I will leave the rest of the word salad alone.


If you really understand and agree, then why do you keep disagreeing?

Then scroll back and read.

The equation describes a numerical quantity that has been named “Functional Complexity.” This generally implies a longer sequence enabling the function, but does not mean the sequence itself is more complex; is could be a chain of “simple” parts.


What JH said. This gives the appearance that you do not understand. I repeat my suggestion: You should try to express what you think it means, in your own words. That will make it easier to see where your understanding goes off the rails.

1 Like

I agree with Joe and not John.

What you wrote does not make sense. Re reading does not fix it :slight_smile:

The equation is a measure of system complexity. Clearly stated in the paper:

Accordingly, we define “functional information,” I ( Ex ), as a measure of system complexity.

Mutation is. Until some S value is reached it is random.

What JH did was make two statements I disagree with. This is independent of Joe’s 1 thru 5 process I completely agree with.

How does the proportion of sequences that function as well as or better than the reference sequence amount to a measure of complexity?

Word salad showing that you understand neither mutation nor selection.

And you can’t support that disagreement with more than a throwaway sentence from the abstract.


So now this is John Harshman’s version of functional information?

System complexity increases with sequence length and the fewer combinations that function. The complexity is in the unique functional information or DNA/protein sequences. The only known mechanism that can produce the level we are observing in the cells is a mind.

Other suggested solutions remain highly speculative and absent a realistic model.

How will selection take place if the S value is zero? If the S value is low enough how will selection make any real difference especially when the protein is part of a protein system. The S value is part of Behe and Lynch’s models. The problem is with a system with high FI genetic entropy is real just based on the mathematics and empirical evidence. The antidote is purifying selection.

You are making assertions that are contradicted by the evidence. When I respond you invoke the “word salad” defense or the “you don’t understand” defense which in my mind is evidence you cannot defend your position.

Why is there no unique sequence for human MYH7? It’s obviously functionally complex.

1 Like

No, it’s a question about the meaning of the definition of FI, which does not mention complexity.

That’s an assertion. Why should complexity increase in that way? As for the rest, you’re just repeating a few comforting mantras to yourself.

It will not, but that has nothing to do with your previous word salad.

“In my mind” is the operative phrase. It’s only in your mind. Consider the alternative possibility that you frequently fail to make sense. You have already agreed that you sometimes express yourself poorly. I suggest that this happens more often and to a greater degree than you realize.


There is no position that JH needs to defend. Your position here, is that lack of understanding makes you right. No.


From the abstract.

Accordingly, we define “functional information,” I ( Ex ), as a measure of system complexity

What is your argument that it is not part of the definition again?

Does the word.
Have the same complexity as the sentence
The apple fell from the tree?
Does the sentence:
The apple fell from the tree
Have the same complexity as
Tha appl feb from th tre.

This is true but you invoke “word salad” and “you dont understand”. too often to be credible. The “you don’t understand” is putting yourself in a position of arguing from authority. “Word salad” assumes that you are immune to misunderstanding a concept due to a cognitive filter.

Answering that would require a complexity metric. Are you proposing one?


Hi Neil
There are several definitions out there but Hazen a Szostak have defined functional information as a measure of system complexity as:

-log2 F[(Ex)] where F Ex is the proportion of functional sequences to total sequence space.

You can use this formula to measure the system complexity of the letter sequences I have written above.

Because the definition is a mathematical formula that does not include a “complexity” parameter or set of parameters that mean “complexity”. What you just quoted is not the definition of FI.

I’m not sure exactly how you measure complexity with words or how you would measure FI with words, for that matter. I would intuitively suppose that “Apple” is less complex than “The apple fell from the tree”, but I have no way to support that assertion. I have no idea whether “The appl feb from th tre” is more or less complex. What does this have to do with FI?

In order to assert that you would have to have a good estimate of how frequently you fail to make sense, which you assuredly do not.

No, that’s not what they say. Nor can you use it (not even the real formula) to measure the complexity (not even the FI) of those letter sequences.


Then I suggest that you work through the details as a way of explaining it to us.


Then why don’t you do just that, and show us your work. I think this should be very illuminating.


And I agree with John, and think he agrees with me.


I think it’ll just tell us what we already know.

1 Like

That’s not the point. This is an opportunity for Bill to demonstrate his understanding to the application and interpretation of FI. Think of it as a learning opportunity.


That would be great. But if you think of it as a learning opportunity for Bill, that isn’t something that Bill does. And it’s not a learning opportunity for anyone else, because we already know that isn’t something that Bill does.


Hi Neal
Let’s assume that we can have 10 misspellings and still understand the sentence and 2 misspellings and still understand that word Apple. Let’s also assume upper case and lower case can be used but no punctuations marks. There are then 53 possible elements a through z upper case and lower case including spacing between words.

If you use google you can obtain an exponent calculator. The system complexity (based on Hazen and Szostak’s paper) of Apple is around 53^5 divided by 2 or around 28 bits. The system complexity of the sentence is 53^28 divided by 10 or around 150 bits.

The equation is based on a combinatorial calculation which I assume you are well versed in.