Can random mutations cause an increase in information in the genome?

Evolution isn't usually studied from an information perspective. The only reason it comes up is because of the creationist objection that only intelligence can create new information, and that therefore evolution is impossible.The error in this creationist assertion is easily pointed out. This is from Message 15: In other words, a mere copying error during cell division can create new information.--Percy

Hi Nosy,I tried to answer this question from the OP: I think the ambiguity in the OP may be due to Garrett's lack of understanding of the Creationist claim and that he's not making original claims regarding information not seen here before. I know he talks about information having meaning, and we can come back to that if he wants to. Interestingly, he is dead on when he says that DNA is not information but only a communication medium, though we don't usually take this particular perspective in discussions here. Anyway, I didn't want to spend too much time on this until he demonstrates he's going to stick with it.--Percy

I suggest you check out Message 13, which describes how information easily increases through replication errors. A corruption of information is not the same as a loss of information. This might seem counterintuitive, but information theory isn't about meaning, it's about information, which it measures as the sum of possible messages that might be sent. It doesn't matter if the messages are meaningless - from an information theory perspective, all messages are without meaning. Or, as Shannon put it in his landmark paper, "Frequently the messages have meaning; that is they refer to or are correlated according to some system with certain physical or conceptual entities. These semantic aspects of communication are irrelevant to the engineering problem."If the change in the beetle's DNA was removal of an allele or even of an entire gene for wings, then the information in the population has declined. But if the change was the addition of an allele or gene that turns off or otherwise prevents growth of wings, then information has increased. Has research uncovered which it is? If by "increase in the specificity" you mean "increase in information", then this would be incorrect. Any mutation which adds an allele to the bacteria population's gene pool is an addition of information. This is a commonplace event during reproduction.--Percy

Not to get into a whole 'nother debate, but the creation/evolution controversy developed out of attempts by the evangelical community to gain representation for their views in science classes by crafting laws and by lobbying boards of education and publishers of textbooks. It was successfully argued that only science should be taught in science class, and so the evangelical community over a period of decades developed increasingly sophisticated science-like scenarios that never mention God or a creator. The necessity for removing God from creation science was recognized early on.So I'm surprised to see you mention God. Almost all IDists argue that the designer is unknown but stress that he is not supernatural. Thus, your approach seems to lose the argument for you outright since it's a concession that you're not doing science but religion.--Percy

I'd like to echo NosyNed's comments in Message 27 and suggest that you restrict yourself to the topic of your thread, the possibility of increases in genetic information.--Percy

But I was on topic. You said this in Message 19: After a paragraph of explanatory background, I said this: You go on to argue in this post: This is the point you're trying to support in this thread, something you haven't yet succeeded in doing. Your argument is that increases of information are not possible through natural processes, and you have been provided several examples of the very thing you deem impossible happening, see Message 20 from Coragyps, Message 21 from crashfrog, and Message 23 from myself. My Message 13 is a more general argument.--Percy

Hi, Randman, haven't you left a whole lot of threads just hanging? Is that just casual phrasing on your part, or did you think that's what I really meant? In Message 19 Garrett argued that ID implies that God put the new information in the genome, not undirected natural processes. God is a religious concept, not a scientific one. The creationist community has struggled long and hard to remove God from all their proposals to legislatures, school boards and textbook publishers so as to give the appearance of being science and not religion. I was just expressing surprise at Garrett conceding the religious foundation of his views so early in thread. Such concessions are sufficient to make clear he's not doing science.Please don't hijack this thread to debate your belief that God is part of the natural universe and is not supernatural. That discussion deserves its own thread.--Percy

I didn't bring up God, Garrett did. Since your initial premise is incorrect, your conclusions are also incorrect.If you have further complaints about my conduct in this thread, please bring them to the [forum=-19] forum.--Percy

I think it would help if you could provide a definition of specified complexity. If you could apply your definition to my example of a population's gene pool that acquires a new allele, that would be helpful.In classical information theory, information is a measure of the number of different messages that can be communicated. My example goes like this. Imagine a population of organisms where a certain gene has eight different alleles. The amount of information for this gene is log₂8=3.Now imagine that one of the newly born organisms possesses a mutation at this gene location that is different from the other 8 alleles, yielding a total of 9 alleles within the population, so the total information is now log₂9=3.12. Since 3.12>3, information in the population has increased.How would you go through the same process of assessing whether new information has been created using specified complexity?--Percy

That is only your interpretation of the sand on the beach. How do you tell the difference between a beach where each grain of sand has been placed precisely according to a certain purpose, and another beach where this was not done?Such questions cannot be answered. In the same way, you cannot tell the difference between a specified and an unspecified genome. The attribution of specified complexity is a subjective opinion, not a mathematical derivation.--Percy

Randomness in terms of information theory has a fairly precise definition. As someone has correctly stated a couple times now, randomness communicates the most information. This is because the degree to which the next bit of information can be accurately predicted is information that can't be communicated because it is information that is already known. In classical information theory, you can't get credit for telling someone something he already knows. An analogy would be the letter "Q" in the english language. If our message set consisted of the words of the english language spelled out using the 26 letters, then anytime a "Q" was received it would be known that the next letter is "U", and so no information is actually communicated by transmitting the "U". The "U" is considered redundant information.A communication medium is used to transmit the messages of a message set. If all messages are equally likely then this is the fully random case, and the receiver of the message knows that the probability of receiving any particular message of the message set is 1/N_m, where N_m is the number of messages in the message set. If the messages are not all equally likely then the calculation of the probability of receiving a given message from the set is more complicated, as is calculating the amount of information transmitted by any single message.--Percy

I'm sensitive to your concern about topic. I'm staying well away from the topic of random mutation.Randomness as it pertains to information theory is, I think, relevant, since it's a key concept of information theory, and is why Ifen has repeated it a couple times. Someone who doesn't understand that the message with the most information content is a random stream of bits doesn't understand information theory. It's a wonderful starting point for introducing information theory concepts.--Percy

Nitpicker! --Percy

Since I last posted it looks like what Nosy feared has come to pass. The discussion has passed from information to randomness.I'd still like to see an answer to the questions that were being asked earlier in the thread about how one assessses specified complexity, particularly how it is quantified. I posted this back in Message 98: --Percy

What I like about my own approach is that evolution happens to populations, not individuals. Individuals have only static amounts of information. In order to consider the same issue of increases of information within a single diploid individual you have to examine the process of meiosis.But I think many people have an easier time thinking in terms of individuals rather than populations, and for this reason I wonder if you could come up with an information example similar to my own, but for an individual rather than a population. We've probably had between 10 to 20 requests for specifics regarding how one measures specified complexity, but we've not had an answer yet, so maybe this will help.--Percy