Here is one reason I don’t think life as we know it is the result of ordinary processes.
In the genetic code, a stop codon (or termination codon) is a nucleotide triplet within messenger RNA that signals a termination of translation. Proteins are based on polypeptides, which are unique sequences of amino acids. Most codons in messenger RNA (from DNA) correspond to the addition of an amino acid to a growing polypeptide chain, which may ultimately become a protein. Stop codons signal the termination of this process by binding release factors, which cause the ribosomal subunits to disassociate, releasing the amino acid chain. While start codons need nearby sequences or initiation factors to start translation, a stop codon alone is sufficient to initiate termination.
Now what happens when there is no stop codon?
A nonstop mutation is a point mutation that occurs within a stop codon. Nonstop mutations cause the continued translation of an mRNA strand into an untranslated region. Most polypeptides resulting from a gene with a nonstop mutation are nonfunctional due to their extreme length.
Nonstop mutations have been linked with several congenital diseases including congenital adrenal hyperplasia, variable anterior segment dysgenesis, and mitochondrial neurogastrointestinal encephalomyopathy.
In other words, it would be bad juju if there are no means of reading of DNA and recognizing where one gene ends and the other begins. In fact, without stop codons, it looks like a DNA-RNA-protein-based life on Earth would be dead.
One could postulate a DNA-RNA-protein-based life that had an alternate stopping mechanism that eventually evolved a stop codon. But that just moves the problem elsewhere in as much as a DNA translation system that contains multiple genes needs a gene delimiting mechanism. A stopping mechanism needs proteins to implement it, but without a stopping mechanism to implement proteins, there is no stopping mechanism. We have then a chicken and egg paradox.
One could postulate proteins arose by a method outside of DNA translation and somehow recruited DNAs and RNAs and then defied all probability and somehow figured out how to code the next generation of proteins using DNAs that just happen to be coding the proteins like the ones that miraculously recruited them. At some point, such scenarios are so out of the ordinary they are not distinguishable from miracles.
Some will argue Darwinian evolution in the origin of life. That’s problematic for at least two reasons.
1. even most evolutionists don’t view the origin of life and origin of the protein translation cycle as part of typical evolutionary theory. So in that sense of the word “evolution”, stop codons didn’t evolve.
2. something dead can’t evolve by Darwinian mechanisms, and if this is an origin of life scenario we’re dealing with dead pools of chemicals.
So Darwinian evolution isn’t a solution. Chemistry isn’t a solution. Physical laws aren’t a solution. Probability isn’t a solution. In fact what we know of physical process would work against evolution of stop codons, not for it. Hence stop codons and life that depend on stop codons did not arise out of ordinary physical processes.
It need not be fatal, if we don’t expect a ‘big bang’ of complete one-pass replication. In a primitive world, proto-replication would not necessarily be a whole-chromosome thing anyway. At the very start, it’s hard to see how it could be.
A polymerase would start until it stalled, with some kind of cap on the growing 3’/2′ end or simply by falling off. But there is no obvious mechanism that would force the polymerase to commence at the 3′ end of the template (if it even had such an end), nor much likelihood that it would get all the way initially, and replicate every last base. A polymerase could start anywhere and stop anywhere. In the first instance, whole-genome replication would probably not be achievable. But ribozyme generation – the same polymerase abstracting shorter segments – would. There is no evolution available to such a system, but there is enzymatic capacity, and hence the possibility for such a linkage unit to help itself. One vital ingredient, I think, would actually be the development of a ligase. Then a stalling polymerase could be turned into a replicating one, and mechanical breaks repaired into the bargain. Then proofreading (getting rid of those stalling caps), and then … !
Which is to say: biology to the rescue, rather than chemistry! In this (highly speculative) viewpoint, nothing is required to promote and maintain homochirality and complementarity beyond RNA’s paired strands, and the greater stability afforded by homochiral/paired linkages. ‘Half-lives’ would increase in the series free monomer -> ssRNA -> dsRNA, enriching the medium in complementary bases and cyclic ribose. Enantiomers would be partitioned, rather than chirally enriched, and the handeness of descendants derived with 50/50 chance from that of the first successful replicator.
Hybridisation is such a powerful tool, and complementary xNA’s so specifically ‘sticky’, I’m surprised it doesn’t get more attention in OoL scenarios.
Yes, I guess the common feature is a priming -OH, which does not have to belong to a nucleotide at all.
Further thoughts on enantiomeric cross-inhibition: the stumbling block here is the impedance provided to non-enzymatic template-directed polymerisation of mixed-chirality nucleotides. The importance of this, though, depends on the primordial model being considered. Bare single-strand oligonucleotides replicating by this means (or at least generating their complement) certainly could not be true replicators in an extracellular milieu because of it – and because of the 2-step process required to reconstitute the sense strand.
However, if there are bare single strands, it is hard to see how they can avoid hybridising with random complementary oligomers, which would act like RNA probes, and tend towards short regions of homochiral complementary dsRNA. Nonenzymatic template-based polymerisation can take place against remaining ssRNA segments, or temporarily melted dsRNA segments, of this longer polymer. Where single, this provides a means, subject to a length cap, of generating a stabilising complement by a means other than hybridisation. Where double, synthesis may be in one direction or both.
The fate of these short fragments depends on their degree of complementarity. They make a single strand double, but may be displaced by a more complementary random oligomer, assuming, as is likely, a significant error rate. Where the strand already was double, they may displace the original complement, or be displaced by it, depending again on relative competition for binding between the 3 or 4 strands involved.
As complete ‘genomes’, there can be no organism-level evolutionary competition, as whole-genome replication is unlikely. But as subgenome fragments, there can be. A clumsy polymerase should be more stereospecific than the nonenzymatic version. Its dsRNA segments would be longer before capping, and it would garner more of local resources – monomers of its preferred chirality – than rivals. Further, there can be LGT. Unbounded, a genome that can manufacture a clumsy polymerase would struggle to keep it to itself, likewise one that manufactures a clumsy ligase. These do not have to arise in the same proto-organism, but getting them together would provide a huge step. The latter could capture the former.
With the usual caveats: I am exploring the implications of the model, not saying what is or was the case.
You make it sound so simple, Allan. 🙂
Oops – sorry! Bit technical … er, velcro thingy binds other velcro thingy, make each other last longer than things that aren’t velcro. Unattached velcro thingies make bits of missing other-velcro-thingy to fill in the gaps. Bits fall off.
Mix it all up and stand ready with a stick to beat back the foul beast that emerges. Voila! La Vie!
Ah, so electricity was Dr. Frankenstein’s mistake. He shoulda used velcro.
Seriously, the trouble I have is that you need to get to the point where replication (and early replication is bound to be imperfect) kicks in and then together with selection you have a mechanism of great power for driving evolution. You are asking a lot of chemistry to get you to that launch platform. The evidence it happened is indisputable (we’re here) but…
What think you on Carter and Wolfenden?
Well, I am trying to get us there, but stepwise. I think you articulate a problem with a prejudice informed by a working knowledge of evolution: we understand very well how ‘Darwinian’ replication leads to tuning via selection, and that is our go-to mechanism. We are in a rush to get that going so we can reap its rewards. But templated replication stalls, and there is no obvious way to replicate an RNA strand which is composed of mixed L and D ribose and contains bases which have no complement. Additionally, to get that strand back, we’d need 2 rounds of replication, by a mechanism without too many errors that manages to copy every base, first to antisense and then back to sense. That’s a big ask, as they say on footie programs.
But I am suggesting a step back and a look at the actual properties of RNA. In order to get a replicatable RNA, we need one made of a consistent ribose monomer, attached to bases that have a complement – which will hydrogen bond purine-pyrimidine when the strands they contain are oriented antiparallel.
One way in which that may be achievable is by hybridisation of random short strands. RNA is a very powerful probe. It finds its complementary sequence with great affinity. You can stick RNAs on a chip and screen for SNPs – single base changes – or you can design RNA primers that will find a region of interest during PCR among billions of base pairs. Antisense transcripts can silence a gene by hybridising with the mRNA. Complementary strands ‘invade’ the homologue during recombination. And so on.
To me, this high affinity provides a potential mechanism by which homochirality and base complementary may be achieved prior to replication – as it must be. Hybrid double strands assist each other’s survival. They pin bases in place such that nicks in the backbone do not result in the components floating off separately. Because homochiral, complementary sequence has the greatest affinity, this ‘selects’ those properties enhancing it from a random pool
There is a limit to the amount of complementary sequence one can expect to find floating around nearby. Where bound, there will be a region of double stranded RNA, but these will be separated by regions of single strand. Nonetheless, the 3′ ends of the double stranded segments may provide start points for extension, initially by nonenzymatic template based replication, gradually removing the single strand.
This, at this stage, is not whole-genome replication, and it is not catalysed by a genome component. But it could be, without this becoming (at this stage) a fully formed replicator. Only partial replication is even possible, I would imagine, initially. But partial replication, for an RNA organism, is essentially transcription – a possible means to crerate functional products. Partial replication is enough to create a ribozyme. And if the ribozyme you create is a polymerase, you have a means to further stabilise the linked unit, by extending the double strand region actively, to the benefit of the polymerase and everything linked. Add a ligase to join a ‘gapped’ monophosphate link, and you have the beginnings of a mechanism of ‘true’ replication.
And then you can have ‘Darwinian’ evolution. But you need to prepare the ground first, and it’s not chemistry I’m invoking, but mainly biology. I’m suggesting a mechanism by which enzymes can come first, then replication, but before anything else, homochirality and complementarity, by self-selection.
1. Just a general point, which I believe I mentioned earlier: speculation only gets us so far, and for origin of life research, its not absence of speculation that’s holding us back, but absence of robust experimental models. I suppose one could look at what happens to chirally mixed and/or non-canonical base-containing RNA strands in a mixture, and specifically see how much you can enrich for base-pairing and homochirality purely by the extra stability of duplexed RNA. That still needs to fit into a model for the origin of RNA oligomers (which we don’t have) and a model for RNA replication, non-enzymatic or via ribozymes (which we are maybe close to, but don’t have yet).
2. Pre-Darwinian events. Beyond everything else, the main advantage of Darwinian evolution is that if A is 1.2x faster growing than B, then the ratio of A/B will grow towards infinite: A takes over. In a more standard chemical system, this isn’t the case, if A is being formed 1.2x faster than B, then the ratio of A/B converges toward 1.2: A’s share is proportional to its advantage. Getting enough homochirality or base-pairing under the latter conditions would be much more difficult. Not, on the face of it, impossible, but I also wouldn’t be surprised if the total effect results in only a slight bias.
3. Carter and Wolfenden. Genetic code numerology is a grand old tradition, and I don’t find anything particularly new in this version. Its pretty clear that there is some grouping of amino acids by physical property in the code, but that its not perfect; and its not clear if this is due to optimization, or reflects history (do isoleucine, leucine, and valine share a middle base because its advantageous to group them, or because they all split from some common aliphatic amino acid synthetase?). Parts of the tRNA sequence and amino acid identity are (obviously) already perfectly correlated, so its not surprising that you can find associations with amino acid properties too. I’m not convinced that their tests against non-canonical amino acids (pyrrolysine and selenocysteine) are proof that they haven’t just overfit their model. And there are enough ways to characterize amino acids that some are bound to fit better than others, so I worry about people (not even intentionally, but as a group) trying properties until one fits. In the end, these correlations don’t help distinguish between the variety of models that are out there for the origin of the code. I’m much more a fan of efforts to reconstruct history before LUCA using phylogenetics (see, for example, Greg Fournier’s recent work), or evolution-informed structural biology (Bokov and Steinberg’s model for the ribosome is a good example).
Blah, if the above seems overly negative, let’s chalk it up to the fact I’m stuck in lab with a head cold and its approaching midnight.
My main point re: 1+2 above is the idea could plausibly help get us to replicating RNA, but its hard to say how much without having more evidence about what needs to happen up and downstream from the “RNA oligos without replication yet” point. I’ll add that folks are looking at forms of hybridization, both in terms of forming oligomers in the first place (Nick Hud) and forming ribozymes out of short oligomers (both Niles Lehman and Phil Holliger have done stuff on this in the last few years).
It’s asking a lot to address the entire sequence! It’s a problem with any scenario – home in on a particular part of the sequence, early or late, and even if you are bang-on – that’s what really happened – someone will legitimately say: “yeah, but how did you get there? And how do you get from there to here?” 🙂 And if you are wrong, of course … well, there are so many trees to bark up.
Just to note that I’m not advocating this mechanism for chiral enrichment, but chiral partitioning. Stable homochiral duplex chains of either predominant orientation form a pair of pools from which one or the other form may discover a ribozyme polymerase. There are advantages to short-segment polymerisation without needing a whole-genome ‘Darwinian’ mechanism – genome fragments themselves are subject to a mini-competition, and mutation, and hence tuning. Once you can polymerise a short segment, it is not a huge step to polymerising the lot. Descendants will inherit the handedness of the ancestor – biology, not chemistry, fixes this. But you probably only have a few weeks or months.
Sure, if I had a lab I’d be doing some work on it!
My own view on the genetic code is that this derives from gradual widening of few assignments or one, and I doubt that co-evolution of codon and amino acid took place, or selective assignment tuning a la Freeland & Hurst (‘our genetic code is 1 in a million!’). I think that the precursor system was noncoding, with priming attachment to a short acceptor stem which grew to become our tRNA. On the way, docking to mRNA helped to drive the peptidyl transferase reaction, providing the beginnings of a coded system. The properties of the code – error tolerance, grouping of like properties – I believe arose from constraints on substitution, not positive selection. It is less disruptive to subdivide a codon group on like property, which naturally becomes the neighbourhood for misreads.
There is a mistake in this thinking. Selection can be a barrier rather than a facilitator to complex interdependence. Selection for a trait after it exists may indicate there would have been selection against that same trait before it exists. This is a subtle equivocation that needs to be pointed out.
If a cell dies today without helicase, that indicates if helicase was life critical in the past it would have led to extinction. Hence one has to suppose helicase was not life critical in the past. Hence one has to argue evolution proceeds without such intense selection pressure as to make a characteristic (like helicase) life critical.
A case could be made that selection would preferably be absent for a characteristic to evolve and then afterward it becomes life critical. But then, even in that scenario (not life critical before it evolves, but life critical after), selection isn’t critical factor in evolution — chance and chemistry are.
I don’t mean to throw cold water on this excellent discussion, but even evolutionist like Stanley Sathe pointed out, the most variable characteristics are those that are not life critical.
So if helicase is not critical, exploring for the helicase protein will be a metabolic load for an organism creating helicase pre-cursors that aren’t used until it finally is able to arrive at a functional helicase and then incorporate it.
The problem of OOL is not replication. Salt-crystals replicate, there are lots of natural replicators. The problem is that life is an extravagant Rube Goldberg replicator. Much of the replication cycle for life has more and more chemically dependent steps for the reaction mechanism to take place.
It’s good experiments happen that show how minimal replication can happen. We learn lots about the properties of RNAs. That’s good.
The problem of life is that we have so many life-critical parts. Selection doesn’t solve the problem of so many life critical parts and steps, it’s a barrier to their evolution.
Re: hybridization. Sure. I think we’re circling each other’s arguments here. In the end my mean expectation is that hybridization might help clean these things up, but probably not that effectively in the absence of some kind of replication mechanism or other form of feedback. But only probably, I’ll certainly admit its possible.
Re: genetic code. That’s close to my own views, though I think there’s a lot of uncertainty, understandably. I’ll reiterate that I really like the kind of work done by Greg Fournier, trying to reconstruct events pre-LUCA from a phylogenetics perspective.
You’re asking about two issues which are for the most part entirely separate. Life-critical traits emerge gradually, as evolution causes an organism to depend more-and-more on an activity (think breathing in air – most modern organisms which do it can’t breath in water anymore, but the first to do it, and some which still exist today, can do both); this result has very little to do with whether selective pressure for the de novo origin of this trait was extremely strong or weak or non-existent. The de novo origin of traits does not require strong selective pressure, it just requires that some behavior that emerges from the “slop” of biological interactions, of which there are many, leads to some selective advantage for an organism, so that gradually that fortuitous interaction is fixed and amplified, maybe separated from other activities by gene duplication etc. I’ll separately add that strong selective pressure does not require that absence of a trait be lethal – it just means that changes in the effectiveness of the trait, even at very weak levels, leads to large changes in growth rate (imagine, for instance, a population of organisms that can’t digest a particular food which is plentiful (citrate in cultures of e coli, maybe?)- ability to digest this food is by no means required for life, but such a trait would be under huge positive selective pressure). I am sure the origin of de novo functions in biology via evolution has been explained to you many, many times, and I don’t expect to do better than others who have tried, so I’ll leave it there.
Using the argument from personal incredulity, it is hard to see any form of ratcheting taking place without feedback from some selective/replicatory process. Is the odd billion years or so and “monkeys on typewriters” sufficient to get us to LUCA?
This is pure nonsense, Sal, if you don’t mind my saying. Selection acts on variation. Organisms (or analogous simpler entities) that acquire (rendomly) variant structures that give some survival (more generally – replicatory) advantage in the immediate environment will tend to proliferate in that environment (as Lenski’s LTEE shows). There is no foresight and no long-term plan. “Selection against that same trait before it exists.”? Good grief!
No Sal, I’m afraid the mistake is in your thinking. Selection acts on what turns up. It is nothing more than a bias in change in allele frequency due to differential reproduction of one type vs another. Sometimes that bias is in favour of the novel allele, sometimes it is against. Sometimes change in environmental or genomic circumstances can lead a previously favourable allele to become disfavoured and vice versa. Selection does not, on the evidence, freeze organisms in a particular genomic configuration. And even if it did, this is no argument against that genomic configuration having arrived through (inter alia) selection.
Once a trait turns up, it can become ‘life critical’ for that lineage. Substitute anything for ‘helicase’ – try ‘wings’ or ‘legs’ or ‘hearts’ – and see if your argument seems so convincing then. Of course, when a selection pressure is removed – see wings – ‘life-criticality’ can diminish. For basic cellular processes, this is unlikely, because further amendment arises in the environment provided by the prior amendment. A course of bricks is not critical when first placed.
I love the “may indicate”. You talk like you are writing a research paper instead of a blog post.
Why don’t you construct an experement that will show one way or the other, that way you can replace your “may indicate” with “did indicate”.
But that’s the trouble with you, you are already an expert, you already know the answers (Darwin beat a dog) so no need to actually dirty your hands with actual work!
Well, this is the nub I guess. How do you replicate segments with no complement on a twisted backbone?
The important point, I think, is that ribozyme formation and replication can be seen as essentially localised and extended versions of the same process. I think there is a case to be made that the former probably must precede the latter, though not by much. Ribozymal models tend to look to self-replicating RNAs – complete single-strand ribozymes with the capacity to make copies – but there are so many problems with this model. I like the hybridising, initially sequence-irrelevant model as it obviates many of these (and, admittedly, imports new ones …)
The field is beset by handwaving. When Larry Moran dismisses the RNA world (for sound reasons, granted), he becomes equally vague when discussing the alternatives. Metabolism kind of starts, there’s a proton gradient and it generates complex carbon molecules and this metabolism replicates and …. RNA kind of takes over … and …
I’m just nailing my colours to the mast! Sometimes people just talk past each other. If RNA World conjours up images of a soup of monomers from comets just polymerising, that’s not what I’m about. If someone wants to call monomer production metabolism, be my guest. But if they start talking of ‘replicating metabolic systems’ somehow becoming replicating nucleic acid systems, I’d like to know a bit more about how they envisage this working. That’s why I prefer to start with the more familiar duplex, with ribozymes fragmentary components of this.
There’s a possibly unintentional suggestion there that LUCA was the first replicator … ?
A limited form of selection can take place without full-genome replication. Evolution is about survival and reproduction. Initially, what I am proposing is a mechanism of differential persistence. Biological molecules do not last for ever, but the time they do last is dependent on the configuration they are in. Free monomers would be expected to last for less time than in a single-strand polymer, which would last for less time than in a duplex polymer. The most stable duplex polymers are homochiral, complementary ones. This would potentially enrich the medium in monomers possessing a complement – a vital precursor to any RNA replication.
Any polymer is a linkage unit, competing against others and against simple degradation. Polymers that are better at persisting will … persist better! Anything they can do to enhance that persistence will help, which includes the possibility of replicating short segments to function as ribozymes, possibly capable of repairing breaks. These segments can mutate – can evolve – in the brief time available to such a unit. Once you have a means of short-chain replication, you are knocking on the door of whole-genome replication. If you have a duplex strand and can copy both strands with reasonable fidelity – kaboom! You are a replicator. Now you can really do some serious tuning, leading eventually to us typing monkeys! 😉
Starting with a genome which is a single ribozyme – ie, starting with replication – has significant problems. Granted that it is a much more powerful mechanism, but you have to get it airborne first.
What do you expect from a dyed-in-the-wool adaptationist and Dawkins devotee?
Your comment makes eminent sense. And the fun is without any evidence of what the key prebiotic environment was, we can speculate freely (within the known properties of matter and energy). I can see that the “development” of prebiotic (prior to self-replication) units might involve some kind of “production” scenario. I’m still drawn to the idea of hydrothermal vents (undersea or hot springs). You have temperatures high enough for chemical reactions, a wealth of minerals, a way to freeze reactions (proximity of very cold water – especially undersea), and every niche empty for the lucky* first replicator.
*There’s the rub!
I’m a fan of Dawkins too, and have adaptationist leanings. His view of a team, with linked futures, is a powerful one – often misunderstood.
You’re missing Allan’s point, which is that the LUCA — the Last Universal Common Ancestor — need not have been the first replicator.
In other words, you don’t need “monkeys on typewriters” to get you to LUCA. They just need to get you to the first replicator.
We can take your word for it?
Getting all the way to LUCA from fortuitous chemistry isn’t necessary – we have good reason to believe LUCA was just one organism within a much more diverse population, and that LUCA itself evolved from simpler life. The evidence for the RNA world suggests that these precursors could have been much simpler. The question of how much simpler is the point at which I’m skeptical of most hand-waiving and even more thorough theorizing, and mostly would just like to see people make simple replicating systems and test them out. I think we’re close to being able to do that, at least for some cases. Hopefully, we can slowly hone in on how simple things can be and still be able to evolve new functions. That won’t get us all the way there, but it will let us pose better questions about what chemistry and luck needs to do for us.
If you can’t think it through for yourself, then yes, I guess you’ll have to take our word for it.
It’s pretty obvious, though. If you can’t visualize it, try drawing it out on paper or a whiteboard.
david, pardon, but our best science resolves to a LUCA. So how can you assert that our best science resolves to something prior to the LUCA?
Your claim that you have good reason to believe that LUCA evolved from simpler life is based on what?
keiths, perhaps Alan is willing to take your word for it, though I fail to see how he could given your past history with him. Given your past history with me, I would be a fool to simply take your word for it.
Empirical. Evidence. Do you have any?
Take my suggestion. Draw a tree on paper and think it through. Be sure to consider various extinction scenarios.
keiths, extinct monkeys can’t type. Think man!
And living Mungkeys can’t think. They just type, man.
So, first, the inferred existence of mitochondrial eve does not imply that she was the only woman around at that point in time, nor that she was the first woman in history. For the same reason, the fact that all life shares a common ancestor does not indicate that LUCA was the only life around when it existed, nor that it was the first: there’s a reason we call it the LAST universal common ancestor.
Much of the positive evidence for life before LUCA is inferred from the fact that although organismal trees converge to LUCA, individual gene trees converge further back than that, especially if you start looking at paralogues within LUCA. I mentioned Fournier’s work earlier, and he has quite a few good papers about this. Here’s one about HGT events from extinct lineages that branched off before LUCA: http://www.biomedcentral.com/1471-2148/15/70
Although more speculative, others have attempted to deconstruct ancient evolutionary events in the ribosome. I mentioned bokov and Steinberg, but heres more recent work that expands on that result: http://m.pnas.org/content/111/28/10251.full
Both papers are open access, so anyone interested should be able to review them.
Mung – do I take it you are happy to accept that LUCA was the common genetic ancestor of all modern life, even if you are skeptical of the evidence of a prior evolutionary history for it?
It’s like we are working out for Mung why he is an ID supporter.
Sorry, just now revisiting this thread.
I don’t even have a problem with the LUCA having a prior evolutionary history. I was just skeptical how much we can know about that history given that our best science resolves to a LUCA.
I don’t have a problem with the idea that there was some population of organisms from which all extant life shares a common origin.
Where is ID relevant to any of this then? Is this “population of organisms” pre LUCA a designed population? Or what?
Selection acts on what exists. The fact it acts on what exists is not evidence it can act on it when it doesn’t exist.
I was pointing out, the fact selection acts against a system missing a life critical part (like say a helicase, a heart value) is evidence against selection being part of the mechanism of evolution, not for it.
You just proved my point. You have to argue against it being life critical in order to evolve! If it is not life critical, then on what basis does one claim it was selectively favored in any differential way to begin with — except by pure conjecture and often in the face of contravening evidence.
We have transgenic bacteria with human insulin created by pharmaceutical companies in their search to produce insulin for diabetes patients. That is a case in point where a bacteria expresses a functional protein for which it has no use for. On what basis will we think the bacteria will somehow evolve the machinery to utilize human insulin?? It can express this protein, but it has dubious benefit and likely some small metabolic cost.
On what basis then do we think something like non-functional helicase precursors will be selected for when the proto-helicase isn’t functional and none of the attendant machinery needed to position it and use it is in place?
The problem is so-called ‘natural selection’ does not seem to agree with natural expectation. Natural expectation would be that there is no differential reproductive advantage to a non-functioning protein, and even if it were functional in the sense that it could hypothetically be used for something, without contextual machinery for it, it won’t lead to differential success — as in human insulins in transgenic bacteria.
The point is, there is not a convincing case FOR selection as a mechanism of creating life critical parts. One has to actually argue how proto parts aren’t selected against or how they, under neutral selection, will still evolve despite combinatorial barriers.
I’d be on your side of the argument if I found satisfying rebuttals to the problems I just posed. I certainly respect your views, but what I said are deal breakers for me in believe selection had much of a role in evolution of life-critical proteins in translation and replication.
There must be some subtle distinction between something ‘turning up’ and that thing ‘existing’ that escapes me.
Has anyone claimed that it can?
I don’t know why the possibility that something can start off not ‘life-critical’, but then become ‘life-critical’, is difficult for you. Selection can promote something to fixation because it enhances reproduction. At some later date, selection can prevent the loss of the system because other parts have built upon it. The selection coefficient of any trait is contextual, not static.
You seem to think that the selection coefficient of a trait is fixed, and therefore, if its loss is fatal now, it must always have been fatal, even on the day it arose. Can you not see what a preposterous caricature of evolution you are arguing against? Do you think the entirety of evolutionary scientists are too thick to have seen your problem, if problem it were? The straw man version is that organisms lacking a presently life-critical trait had a selection coefficient of 1 the day it arose. Suddenly, not having it became lethal, where the day before it was fine. Obviously that’s ridiculous, but there are simple and obvious ways in which a selection coefficient can become 1 without starting there.
X will not evolve therefore NOTHING evolves? That’s a bogus argument.
Nobody proposes that a nonfunctional component will be subject to positive selection. Still, a poorly functional component is better than nothing.
Therefore … ? Come on, you’re a smart guy, start filling in the gaps. Wonder to yourself why evolutionary biologists are not stumped by your conundrum.
In the land of the non-helicases, crap helicase is king. In the land of excellent helicases, crap helicase is maladaptive.
I can’t spoon feed you or lead you to the answers. But to me, you are arguing against straw men – multiple ones. You pick geology, physics, biochemistry, biology … all wrong, according to you, at a very fundamental level. But each time I read your arguments, I see clear errors in them. I try and address them as best I can, but I’m dubious of your stance that you only wish to see a convincing argument. I think that something in your mindset is actively opposing objective assessment. There seems to be a fundamental difficulty within Creationists to even think in evolutionary terms – to properly understand the paradigm they oppose.
I’d find it more believable that helicases can evolve when none exists if we had an experimental demonstration. Same with critical aaRS genes. We all know what will happen if knocked these out in present day creatures.
In absence of that, on what basis will a non-functioning proto-helicase evolve via differential reproductive success? You can say you believe it will, and I can accept you believe it. But what would count as evidence to you that selection was involved in making the proto-helicase into a functioning helicase? If you speculate that is the case, that’s fine, but perhaps it should just be labeled as such — a speculation.
A few comments ago, the assertion was made that once replication happens Darwinian evolution takes over. I was pointing out, one can make that assertion, but there is little in the way of proof — by proof I mean relative to other scientific theories like Celestial Mechanics or Electromagnetism or even the octet rule of thumb in chemistry.
Yes, and we all know what would happen if we take your heart out too. So what? Hardly seems fatal to the possibility of evolution to me. But of course such a stance nicely insulates you from EVER accepting evolution. Because one would have to obtain the actual creatures and subject them to the exact same selective pressures, before your very eyes. Clearly impossible. You can deny everything but that which is in plain sight.
Meanwhile, you have no problem with the idea of a Deity issuing some fiat and the entre shebang arises in a day … ! Were they haploid or diploid? Feet or ears first?
The proto helicase is a functioning helicase. Just a poor one. It gets better because it improves the replication of organisms that have it. Sure, I have no direct proof – the tired old ‘were you there?’ gambit. So yeah, if you think the speculation unreasonable, don’t buy it. No-one is forcing you. Is the path to Heaven more sure if evolution is denied?
Apples and oranges. Evolution is about matters of history (although equally, it is a fact that differential replication leads to evolution). There are fundamental principles, and they can be shown to work. Their operation can also be reasonably be inferred from molecular phylogeny in many cases (though not for much pre-LUCA, which is what we are talking about for these cellular universals). But you seem to want any evolutionary incident to fall into some kind of scientific inevitability, else it did not happen. It doesn’t, precisely because evolution is unfolding of history. It’s like demanding that a historian re-enact the Mongol hordes, or you won’t accept their existence. It’s a short step to Last-Thursdayism: there is NO history, because we cannot reconsitute every last incident that led to the current positions of all the humans on the planet. Therefore they just popped up? Or perhaps some history is simply cryptic. Why can evolution not be like that?
To clarify, I meant something that doesn’t function as a helicase. If you have a better term, I’m happy to use it for future reference. Yet-to-be helicase polypeptide. Sorry for the confusion, I hope this clarifies.
Back up a bit. Let’s look at how we attempt to establish selection strength and coefficients.
In the present day, we can expose bacteria to antibiotics. Traits that are strongly selected for are obvious. It’s is relatively straight forward to calculate selection strength in the case of antibiotic or pesticide resistance. But this is selection for EXISTING traits, not traits that are yet to possibly exist. How do we even affix S-coefficients to yet-to-exist traits?
We supposedly calculate selection strength by the amount of conservation. But this sort of falls apart when we knockout deeply conserved regions in mice and there is no noticeable effect on reproductive fitness. At best it shows we just don’t know how much selection acted on the ultra-conserved regions, at worst it shows a theory was just concocted in absence of actual experiment. But even in that case, it is selection for EXISTING traits, not yet-to-be existent traits.
So let’s apply just a reasonable level of skepticism to the claim that goes something like this: “once a replicator exists then Darwinian evolution can take place and add more complexity.”
How can that inference be made outside of circularly reasoned conjectures like was made for ultra conserved regions that can get knocked out? Or worse how do we handle the fact selection causes observed reductive evolution, but we are hard pressed to see as much new complexity emerge versus the amount that selection can easily destroy. As Allen Orr said, selection can destroy design and had some choice words for Daniell Dennett. See:
If we don’t have empirical or theoretical methods to make the inferences on the S-coefficients for yet-to-exist traits, then isn’t it a little imprecise to invoke selection as a mechanism to evolve traits that did not exist at one time.
Even hypothetically, we might have to say, the S-coefficient was this or that for this long. We can’t establish it empirically, but that would be a necessary trajectory. But that would at least be a start into the inquiry.
Ok then, now that we have a hypothetical requisite S-ceffiecient acting for this long in this way, argue from chemistry and that such an S-coefficient is justified when the trait or polypeptide is yet-to-be functional.
Chemical theory leads to natural expectation of what happens, and chemicals are under no obligation to generate S-coefficients toward yet-to-exist traits. A credible theory would be connecting chemical principles to generation of S-coefficients, but in the case of origin of protein translation and cellular replication, it is just assumed sufficient S-coefficients are there and the chemical inevitabilities force these S-coefficients to emerge for yet-to-exist traits.
Independent of how dumb or looney someone thinks of me, the problem still remains how selection pressures can be reasonably inferred to exist for yet-to-be-existent traits. If it cannot be quantified, it is premature at best to say Darwinian evolution takes over and builds complexity once a primitive replication schema is in place.
I’m ranked #81 in the encyclopedia of American Loons, I beat out Sean Hannity at #168 and even William Dembski at #103, but I held no candle to Glen Beck at #17. 🙂
Not really. I don’t see any strong reason for thinking the first enzyme with helicase function was a peptide. Nor does it require a dedicated helicase. A polymerase could possess simultaneous helicase activity. The role of a helicase is to separate xNA strands, but a polymerase could do that itself, albeit less effectively.
We don’t. You keep saying people do, I say they don’t … how long we gonna keep this up for?
That’s a retrospective measure, not relevant to the spread of a novel allele. Clearly, when the allele is not even fixed in a single species, it is not conserved in any sense.
It is conservation of existing traits. You are lumping positive and negative selection into one bucket.
Complexity is not a necessary result of the ‘Darwinian’ (sic) process. Once a varying replicator exists then ‘Darwinian (sic) evolution can act. This is definitional, not a claim. You may apply skepticism as to whether that ever happens in nature, though that skepticism would be a little on the hyper side, for someone who has just invoked antibiotic resistance.
Allen Orr the anti-evolutionist? No, the Allen Orr you would on every other matter argue against like you are arguing against me. So why bring him in on your side? Arguments from authority don’t wash with me I’m afraid.
But note the operative word ‘can’. It’s a relative of the ‘some mutations are bad therefore all are’ argument.
Selection acts on what turns up. I believe I may have mentioned this. But when something useful turns up, selection concentrates it. When a useful variant on that turns up, selection concentrates that. Gradually, it is tuned. I don’t know what you think would be doing that tuning, if not selection, after the fact of the mutation arising.
Historic selection coefficients that far in deep time are not accessible. But this has no bearing on the logical argument. You wish to know how a helicase could possibly arise in an organism that does not possess it, given that helicase is ‘life-critical’ now. I have sketched a mechanism. Now you adopt the classic Creationist gambit: Prove It. Do it in a lab. Obviously, this is something that (if it happened) happened in the wild over many generations. It is not amenable to demonstration. But the Creationist argument is silly – because you can’t demonstrate that something is possible, it is thefeore impossible and God musta dunnit (in a day, to several trillion trillion cells at once, yet). If it makes you happy and gets you into Heaven, great, but I find it hopelessly unsatisfactory as a hypothesis.
Well, they aren’t, for the nth time. An organism successfully reproducing without a helicase is under no obligation to evolve one. But if one does, and it confers a reproductive advantage … surely I don’t need to join every last bloody dot? 🙂
Good grief, antibiotic resistance is referring to traits that exist.
The question was about evolving traits that don’t yet exist. If you say that can’t be done, then what point is there saying, “selection evolved a trait that didn’t exist at one time but now does?” For all you know it evolved despite selection pressure against it forming.
I’ll agree that Natural Selection preserves a functional trait if it already exists, I disagree it will evolve it before it exists in the first place, at the very least, such claims are not proven.
Well maybe you can connect the dots for evolutionists who make the following unsupported claims that you can easily google:
Selection preserves adaptations. The fact that it preserves them is no proof whatsoever it evolved them — like the Helicase. As I said, knock it out, the thing dies. So how then would such a knock out experiment prove selection evolved it before it existed?
If it doesn’t, then we should stop invoking selection as the primary mechanism of evolving traits that now exist but didn’t exist at one time.
Retrospective speculation. “Measure” gives the idea credibility it doesn’t deserve. The fact the ultra conserved DNA got knocked out without noticeable effect demonstrates the claim of “measure” is dubious. Falsified speculation would be a better phrase.
Word-gaming. One can measure the degree of conservation of various regions of the genome. One can infer that this represents varying degrees of selection. Nonetheless, this is a smokescreen against noting my actual point: a novel allele – or any allele at less than 100% frequency in the population – is not conserved, but may still be under selection. Conservation and selection are related, not synonymous. It’s that distinction between positive and negative selection again. Uphill and downhill aren’t the same thing, nor are concentration and dilution.
1) It’s the some = all fallacy again. Clearly, there is some ultra-conserved DNA which cannot be knocked out without effect – eg the helicase gene, as you yourself have noted.
2) From population genetics, one can compute that a selection coefficient making a difference of as little as 1 birth in 1000 is sufficient to conserve a sequence. Simply observing that ‘the mice seem OK’ is not sufficient to demonstrate that the knockout is neutral, and selection was not involved in conservation. Powerful stuff, exponentiation.
??? How many more times? No-one is proposing that selection acts on non-existent helicases. When people talk of ‘selection for’, they are referring to opportunities. If utilising a new food source would produce more offspring, any mutant that does help ultilise that source will prosper, but only if it arises. If improving your replication process would produce more offspring, any mutant that does help improve it will prosper, but only if it arises.
Mutation/recombination generates a new trait – which, in its first incarnation, may be a very weak version. Selection (differential reproductive success) causes carriers to outcompete non-carriers. Further mutation/recombination generates novel versions which perform better against the background of the weak version. Selection causes carriers to outcompete non-carriers. And so on.
How would that not be ‘selection evolving a trait’? The path to improved function incoporated rounds of selection, and would not have got there otherwise. Of course there was mutation as well.
My turn to say “good grief”! It probably didn’t. If there is selection pressure against something happening, it probably won’t happen. This leaves a fairly wide arena of things-that-can-happen, unless you wish to take the stance that nothing can happen – that there is no such thing as a beneficial novel allele. .
NO-ONE is claiming that selection operates on non-existent traits. I’ve said it, I’ve said I’ve said it as many times as I’ve said it …
Take your heart out and you die. Have you thereby proved that hearts cannot arise by evolution?
Take a helicase knockout (H-). Its selection coefficient is 1 – it is lethal. Go back in time to an organism with no helicase. Can we reasonably infer that, because disabling helicase is lethal now, its selection coefficient was always 1? What reason can you give to dismiss the perfectly plausible hypothesis that, when helicase arose, the H- selection coefficient was not 1? We don’t need to know what it was, we merely need to deal with the possibility that it was not 1.
Whichever locus generates the H+ phenotype, a moment before H+ arises, we may propose in the simplest case that all instances of that locus have a selection coefficient of 0 relative to each other. Now we have a novel allele, H+. Does evolution propose that all H- instantly die – their selection coefficients instantly flip to 1? No – it proposes that, if selection was involved, H+ simply increased its exponent, its rate of increase, relative to wild type. But many years, many rounds of selection and dependent processes later, it becomes embedded, and knockouts are now lethal. The coefficient can become 1. Its present value has no relation to its historic values.
Can you rule this possibility out? Or do you think it is my job to rule it in? And if I do it for helicase, what’s my next task?
Agreed. At least one thing we agree with. 🙂
Agreed on that claim, have always agreed on that claim.
But I disagree on the consequence of the claim. If S-coefficient demonstrated for Helicase today, it says nothing of selection’s role in evolving it before it existed. Evolutionist say “it evolved an adaptation in order to survive” whereas a far more accurate depiction would say, “it survived because of a providential adaptation”.
If “it evolved an adaptation in order to survive” is hypothetically true, the claim would be untestable. Even today, in Lenski’s lab can we calculate the selection coefficients acting on precursors to future traits a million years from now?
That’s why we disagree on this stuff but not other questions, the reliance on non-observables as an explanation.
That said, we can demonstrate absence of the helicase is lethal. In that respect, the everything-first model will work theoretically if one will, like Koonin, accept some mechanism to bridge astronomically remote odds. If I weren’t a creationist, I’d go with Koonin’s explanation — at least he can get the numbers to work.
Chuckle. What happened to “inference to best explanation” and “following the evidence where it leads?”
Could have fooled me. See later, where you explicitly (if unknowingly) deny it again. I’ll say KABOOM! when I get to it.
Selection had NO role before it existed at least at some minimal level …. I’m going to have that put on a stamp, save time.
The present selection coefficient for absence of helicase gives a rate-of-increase of zero – lethality. It is an entirely realistic possibility that, in the past, it was not lethal. If a large chunk of modern replication evolved in the presence of helicase, you stop the whole thing working if you knock it out. This does not mean that the day it arose, everything was similarly dependent. See my earlier comment about bricks.
Detriment and benefit are measured by relative rates of increase. If H- is the only game in town, there is no relative competitor, and no helicase-dependent superstructure. When a helicase arises, if that is beneficial, H- suddenly becomes detrimental, though it could hardly become lethal overnight. All that is required is that H+ have a rate of increase greater than H- .
The latter tends to be how it is put anyway, other than in loose talk.
Of course not. Can you predict which genes of yours are going to be still in the population in a million years time? Does that inability invalidate genetics, or the fact that you are a mosaic of your ancestors? But anyway, nothing evolves an adaptation in order to survive. The claim is not made, other than in loose (or maybe Shapiro) talk, so it does not matter if it is testable (a Creationist grumbling about testable propositions, already! 🙂 ). Some species adapt, some go extinct, some just rumble on.
KABOOM! – proof that you DON’T agree that present selection coefficients are no guide to past ones, contra your earlier insistence that you agree. You are insisting that, if helicase absence has a present coefficient of 1 (is lethal), that has some relevance to the past, and it was (probably) always lethal.
You don’t even know what the probabilities are, so I don’t see you need a mechanism to boost the number of trials available to accommodate them. Certainly, I don’t propose an ‘everything’-first model (whatever ‘everything’ might be). There are substantial thermodynamic difficulties in getting ‘everything’ into place before allowing any flow down the gradients of electronegativity.
It is amusing that you find a gradual accumulation of improvements to be impossible, and demand detailed history of every component, yet some Deity waggles his beard and it all just … kinda … happens!