A deal’s a deal. Starting tomorrow, I expect you to start defending special creation.
1 Like
Hi Joshua
First, thanks for the kind words. I think you have represented my position very well. I am interested in pursuing the chimp and man discussion with a goal to better understand the two competing hypothesis.
[quote]We have looked at this. And it makes a ton of sense in light of neutral theory and common descent. This is exactly the data that convinced me of the scientific strength of evolutionary theory. In the details, in the math, evolution makes rigorous sense of this.
Of course Special Creation is always a valid hypothesis, but we need a clear model of it to test against the data. Currently we have no model for special creation that makes sense of the data.[/quote]
Can you help me with the math here? How many mutations are you assuming and what model are you using to determine how long it would take to get those mutations fixed in the population?
I think the splicing data is a case for common descent or special creation depending on further investigation. Where did these splicing codes come from? I think this will take time because I don’t think we understand exactly how alternative splicing is translated from DNA to the spliceosome.
Is it possible to go from a chimp to man with random changes followed by selection changing splicing codes. This could be confirmed by experiment once we know how to modify the splicing codes. Changing exon sequences could be very harmful to the animal.
One other issue is the titan muscle protein differences between chimps and man. The titan protein is over 30000 amino acids in both species but is about 8% shorter in chimps. This change being by mutation is highly suspect. The titan protein is highly mutation sensitive and its length would likely correspond to a change in myosin and actin in order for the muscles to operate properly. Again this could be confirmed by experiment.
The interesting new idea is that alternative splicing played a large role in the evolutionary story.
1 Like
Can you explain why you think Axe’s methodology does not lead to an accurate calculation of how much DNA must be worked through to evolve through mutation the enzyme sub unit that he modified.
I can see a case for starting near the edge (temperature sensitive) of the functioning space to see the likelihood the mutations would move you toward or away from function.
I agree he is not directly measuring binding but ultimately the bacteria’s ability to survive in a penicillin environment is where the rubber meets the road.
And if you had been following my meanderings on these boards … you will have noticed that I am one of the first to remind newcomers that criticizing BioLogos for evolutionary scenarios that don’t include God’s intervention or involvement is like “going to a Boy Scout convention and criticizing them for not wearing skirts!”
I can’t be 100% positive, because I don’t believe we have surveyed the board participants (let alone BioLogos members), but I think there must be plenty of BioLogos supporters who are Christians and who are content with the idea that God intervened on the difficult parts of Evolution!
These are different words. Everyone connected to BioLogos (at least in an official way) thinks God is involved with evolution. But you’ll find a variety of opinions about whether God intervened in natural history. Generally speaking, BioLogos folks believe that the very nature of natural processes is that God’s purposes unfold without a need for his constant (or even occasional) intervention. But again, that’s very different from involvement. In fact, one of the main goals of BioLogos is to help people see the difference between these two words, as applied to God’s interaction with nature.
If someone was to believe that evolution happened with zero involvement by God, they would fall outside of the BioLogos belief statement, so we would not have any official connection with them.
5 Likes
I was hoping you would introduce the possible ways of making these distinctions.
My personal view is “In for a Penny … In for a Pound” - - meaning once you assert that God is involved in Evolution, even if it is only to set up the rules of Evolution, God’s intervention is implicit all the way through it.
For example - - the Asteroid that Killed the Dinosaurs . . .
Did God extend his phantom spirit hand and create that asteroid out of nothing, and then hurl it right towards Earth at the right time and place?
OR…
Did God design all of Cosmic creation in exactly the right way so that the Asteroid would be formed by the collision of two worlds… and be sent in just the right vector to arrive at Earth at the scheduled time?
So is there really a difference in the two when it comes to the role of God?
Our good friend and associate @Eddie liked to use the term “Front Loaded” as a reference to God setting all Creation in a Precise and Specific Order to accomplish all the various things God wanted and when he wanted them. I don’t think Eddie liked to state specifically that he endorsed a “Front Loaded” view of the Cosmos, but I was not shy about it at all. For example, I saw Cosmic rays as a perfectly legitimate way for God to arrange to mutate a DNA molecule here … or an RNA molecule there. And Biblical historians are quick to point to God’s use of weather as just another way God would use natural law to accomplish His ends, rather than think God “whipped up a supernatural rain storm” whenever he needed one.
In any case, the BioLogos mission page is pretty open-ended about which way God might be “involved” in Evolution. The Mission Statement certainly doesn’t preclude Miracles (right?) … and yet it also accepts God’s work through natural/lawful means.
[quote=“Billcole, post:67, topic:9418”]
Can you explain why you think Axe’s methodology does not lead to an accurate calculation of how much DNA must be worked through to evolve through mutation the enzyme sub unit that he modified. [/quote]
I just did!
Axe claims that his data apply to wild-type proteins, but he only studied one mutant protein, carefully preselected to already be on the edge of stability.
I can’t. Maybe you can explain?
[quote]I agree he is not directly measuring binding but ultimately the bacteria’s ability to survive in a penicillin environment is where the rubber meets the road.
[/quote]No, if one is going to study enzymatic activity, one needs to assay enzymatic activity. It’s incredibly lazy, Bill.
Neither statement is relevant, as the searches aren’t random and we know that very little of space is occupied by functional proteins.
@Billcole we do know how to change the splicing codes, and there is a large body of work on this. We have already done experiments on this. There was even a nobel prize in 1994 (over twenty years ago) granted on this: Phillip Allen Sharp - Wikipedia
Tissue specific splicing is regulating largely by a small number of bases (sometimes a single nucleotide) in messenger RNA. http://www.nature.com.libproxy.wustl.edu/nature/journal/v465/n7294/full/nature09000.html and we even have success in predicting what the specific impact of muations are on alternative splcing Model-based detection of alternative splicing signals | Bioinformatics | Oxford Academic.
Moreover, the necessary (and nearly sufficient) DNA signals for splicing are very short and therefore very easy to evolve by chance. http://www.imgt.org/IMGTeducation/Aide-memoire/_UK/splicing/ Alternative splicing and evolution: diversification, exon definition and function | Nature Reviews Genetics
This is the reason why splicing changes are so common. Because just a one or two nucleotide change can alter splicing sites, variants (relatively) commonly arise. Because a (relatively) high proportion of alternately spliced genes are well tolerated because they retain the bulk of the original function. This means, just like synonymous mutations, they arise frequently are are not eliminated by purifying selection. This is exactly the basis from which neutral theory starts.
Because splicing changes are common and usually neutral, their bulk statistics serve as a signature of a lineages history, more than a signature of shared function. That is how common descent and neutral theory makes sense of these questions:
Notice, if alternative splicing was more correlated with function, we would observe that tissue (e.g. livers) from different species would cluster together. That is not what we see. Instead, just as neutral theory predicts for mutations with these characteristics (common and usually neutral), we see them clustering by phylogeny (e.,g. shared history/ancestry), even in differently functional tissues.
This is not to deny the functional importance of some splicing programs. Some of them are functional important. This also is what enables evolution by splicing. Splicing is a “safe” way of introducing a large amount of variation into a population that can later be selected. However, the vast majority seem to be the product of neutral drift, functioning as molecular clock.
Yes, there is a large space of possible AS, but we see that it correlates most with common descent, and not with function.
I’m going to use some rough numbers here and approximate formulas to make this very easy to understand. As a result, the numbers are not going to match up perfectly. If one added “error bars” to my estimates, you would see very good concordance between this math and the observations.
So we diverged from a common ancestor with chimpanzees about 6 million years ago (mya), and we have about 30 million point mutations between us and chimps. That means we need to explain how human ancestors evolved 15 million mutations (one half of 30 million) in 6 million years.
So, from population genetics (and verifiable with simulations) we know that populations “fix” exactly the same number of mutations as the average individual’s mutation rate per generation. Fixation (population genetics) - Wikipedia and Coalescent_theory.
We can experimentally measure the per generation mutation rate in humans and chimpanzees by sequencing parents and their children. Variation in genome-wide mutation rates within and between human families | Nature Genetics Genome-wide patterns and properties of de novo mutations in humans | Nature Genetics. There is some debate here (because there is some experimental error and this is hard to do) but we get estimates of about 50 to 100 mutations per generation.
So how many generations were there in the last 6 million years? Generation times now are about 25 to 30 years in the developed world, but much shorter in the undeveloped world. Over ancient pre-history, we think generation times are probably about 15 years. So there have been about 6 million / 15 = 400,000 generations since we diverged from chimps.
Like I said, so we have a very rough estimate of 20 million fixed mutations (50 mutations / generation * 400K generations), which is very close to the 15 million we calculate from the divergence data. So, basically, this tells us that neutral drift (neutral theory) explains quantitatively why humans and chimps are so close in genetically. The same computation show why mice and rats are 10x’s farther apart.
So how many of these mutations are functionally important? At least functionally important enough to be selected? We do not know for sure, but we think probably just thousands were important.
How many of these specific mutations were precisely necessary? We have no idea. In biology, frequently very different mutations can have the same functional effect. We have no way of counting up all the ways that theoretically could have had the required functional effect. But we do know that there are multiple genetic “solutions” to an problem in biology.
And how many mutations did our ancestors explore in the last 6 millions of years? Probably trillions. We need a population estimate for this. A very low average population estimate would be 20,000 individuals. And 20K individuals * 400K generations * 50 mutations / generation gives us 0.4 trillion. But this doesn’t include all the miscarriages from lethal mutations, and the fact that this is just a low population estimate. So the real number is much higher.
Keep in mind, there are just 3 billion nucleotides in the human genome. So just about every point mutation could been searched several times over by our ancestors over the last 6 million years. And just a single nucleotide change is enough to alter splicing in a protein.
So we had over a trillion mutational chances to hit on a few thousand of the correct mutations to encode a human. And, likely, none of these mutations were specifically required, because other solutions were possible. So this looks very mathematically possible.
Of course, none of this is an argument that God did NOT directly inspire any mutations by first cause. Maybe He did. But it is very difficult to establish that even a single mutation has important functional impact, let alone reliably pick out the important thousands of mutations out of the 30 million differences we see. A statistical signal for God’s involvement here, giving the data, seems well beyond the reach of any human science.
So, I hope that math makes sense.
And it does have the role for two reasons.
-
They provide an example of a “molecular clock” that tracks with phylogeny Molecular clock - Wikipedia
-
They provide an easy path to largely safe variation that can be later selected in rare cases. These rare cases, though, are exactly how new functions and structures evolve.
So yes, alternative splicing is very important to the evolutionary story.
4 Likes
Eddie,
The fact that the variation is random does not make the entire search simply random, in the same way that the randomness of the deal does not mean that the winners of a poker game are random.
Neither search begins from a random start, so both are highly constrained.
2 Likes
It’s not a new idea.
More importantly, there’s a lot of junk in splicing. We have very good data for that from many, many different genes.
2 Likes
Bill’s original claim was that the vastness of sequence space precludes the possibility of any randomized process finding a new protein binding event. The immune system example shows that this is not the case, since if the sequence space issue was the problem Bill was claiming it to be, then forming antibodies with high affinity for an antigen would not be possible. That’s the point I was making - turning the discussion to the formation of the immune system is a separate issue. I also provided an example of a non-immune system, novel protein binding event.
On that point, however, there is good evidence that the vertebrate adaptive immune system did arise through mutation and selection. Did you know, Eddie, that the enzymes responsible for making the randomized rearrangements of the gene segments that make up antibodies are modified transposases? These human enzymes, and the sequences they work on, were co-opted from mobile genetic elements. So, yes, there is good evidence that the vertebrate immune system is the result of stochastic processes involving mutation (and subsequent, non-random selection, of course).
If you, or anyone else, is interested, google for “Rag 1 transposase”. I’ll see if I can find a decent summary accessible to a lay audience, and if so, I’ll edit this post to include a link.
Edit: I didn’t find a decent lay summary, but you can read the abstract of this paper to see what I’m getting at. For the truly curious, this paper from 2015 updates the previously linked paper.
4 Likes
Hi Dennis
Are you claiming that the immune system is purely a random search process including somatic hyper mutation?
Thanks for your comments Joshua, as they provide a more realistic view of the subject matter. At the risk of sounding argumentative (this is not my intention), falsification of a model entails more than showing that it may be modified by additional data. Thus, the theory underpinning the CD of say, humans, has not been developed in competition with a theory that seeks to show otherwise (perhaps cousins with octopus, or perhaps due to a novel event, or perhaps …?).
This is why I mentioned the “tree of life”, as this has been taught as a symbolic representation of the foundations of evolution - as variety and selection account for the production of all species, and their development/evolution illustrated by the tree of life. If we add ad-mixing and horizontal gene transfer, how is this theory changed, modified, of whatever. I understand other views have been proposed, such as a “bush of life” or a “forest of life” which may provide representations of alternate outlooks, but I am not aware of detailed theoretical work that develops such ideas into competing theories to the current biology paradigm.