What does this have to do with anything?
You wrote this:
Well, we don’t see such mutations happening right in front of our eyes.
Let’s say we went back to the common ancestor of humans and chimps. If you watched each and every mutation happen in the human lineage and watched them accumulate you would not categorize a single one of them as being an increase in information. In fact, I’m pretty sure we could say the same for all of evolution.
This is because there is no definition for an increase in information in ID/creationist circles. There are the fake definitions that are given and then quickly withdrawn once it is shown that these increases actually exist.
Back to the original question. I like to ask to seriously look to the scheme which describes the genetic code. The scheme is extremely well structured. This is not because the drawing is done so. The structure is in the code itself. The challenge is to imagine how such a code could have evolved. I calculated myself, and this code much is more resistant to mutations than a random code would be.
This is a nice example of Maxwell’s demon.
Actually, I think you suffer from Morton’s Demon:
“Morton’s demon was a demon who sat at the gate of my sensory input apparatus and if and when he saw supportive evidence coming in, he opened the gate. But if he saw contradictory data coming in, he closed the gate. In this way, the demon allowed me to believe that I was right and to avoid any nasty contradictory data.” (Courtesy of the late @gbob.)
2 Likes
A nice example of mortons demon.
Your seeming denial of natural selection removing deleterious mutations would be an example of Morton’s Demon.
1 Like
Natural selection’s removal of deleterious mutations is well documented. Only Morton’s Demon can account for ignoring that fact.
I tried to explain several times that there is no natural selection in non coding sequences. Several times, but T_aquaticus’ Morton’s demon is sufficiently skipping that fact.
Everyone is missing things here. @T_aquaticus missed that you were talking about mutations that occur before the sequence becomes coding. You’re missing the fact (and it is a fact) that completely random peptides are sometimes both functional and beneficial, regardless of the mutational path taken to get to the random sequence.
ETA: Also, there’s lots of selection on noncoding sequence – most functional sequence is noncoding. Even nonfunctional sequence has weak selection acting on it, since some mutations will represent deleterious gain of function changes.
1 Like
Sure there is. There are non-coding sequences that are functional, and deleterious mutations in those sequences are selected against. Examples of non-coding functional sequences are transfer RNA, ribosomal RNA, micro RNA, and transcription factor binding sites. If a mutation causes a previously non-functional sequence to produce a deleterious function then that mutation will be selected against.
1 Like
Agree. It is easy to misunderstand each other. Indeed there is a difference between the definition of ‘coding’ and of ‘functional’. And I ment functional. Since we are talking of information and that is linked to function. Which is broader than coding for proteins. So my claim is that there is no selection for function in sequences that are not functional.
Today is world tuberculosis day. I joined a scientific webinar on mycobacterium tuberculosis. It was presented that M tuberculosis evolved from non tuberculosis mycobacteria by: gene loss and horizontal gene transfer. That’s it.
I haven’t followed this conversation closely, so I might be repeating things other people have said, but I thought I’d throw in my two cents, as a linguist.
The problem with asserting that DNA is like a language because it is like a code is that linguists do not use the code model of communication to describe how language works anymore. They use an inference-based model. The success of AI in being able to generate natural sounding language and natural sounding translations from one language to another came when they just fed computers massive amounts of real texts and they learned how to copy based on probabilities, not based on treating words like information units or grammar like rules. Linguists would not say that words “contain” information anymore, they would say words trigger inferences, and those inferences are dependent on a host of things that are not “encoded” in the semantics of the words, such as the physical context, links between conceptual domains, learned cultural knowledge about the world, internalized conceptual frames and scripts that are conventional in a speech community, internalized prototypes, implicit cognitive models, and power dynamics between speakers.
The main way languages (as idealized abstract constructs that represent the totality of what a population of fluent speakers can do with utterances) relate to biology and evolution is that languages evolve within populations of speakers over time and can be grouped into nested hierarchies of genetic relationships. So we speak of language family trees and extinct languages and the transmission of specific features of a language from generation to generation. So the real functional analogy is not between DNA and words of a specific language, but between the abstract concept of population genetics and the abstract concept of language change.
5 Likes
I feel like Biologos ought to be some sort of accredited agency that can bestow a degree on anyone who has read at least ten of your posts, Christy. I’m indulging in hyperbole I know - but seriously - I always feel a whole lot smarter after reading your stuff.
Somebody should go through this entire forum and gather all your posts together. Enough material to publish a great book.
4 Likes
Aw, thanks, Merv. Someone asked me once if I would ever do a blog or a Substack and I said, nope, if you want to be privy to all my external processing and semi-coherent brain dumps, you just have to hang out with me on the BioLogos Forum. 
(And I learn a lot here too.)
2 Likes
That goes without saying since natural selection can only see what affects fitness, and you need something that has function in order to even apply.
If a mutation in a previously non-functional sequence produces a beneficial function then it will be selected for. If a mutation in a previously non-functional sequence produces a negative function then it will be selected against. A mutation can also produce a function that is neutral, and it will fix at random. A mutation can also not produce a function or impact fitness, and that mutation will fix at random.
1 Like
If we compare the human and chimp genomes would you say that the genetic differences between the genomes are gene losses, and that human physiology and function is the result of gene loss as compared to other apes?
1 Like
This topic was automatically closed 6 days after the last reply. New replies are no longer allowed.